Sulphur poisoning fuels

Solid oxide fuel cellsoperateatvery high temperatures, around 1,000°C. High temperature operation removes the need for precious-metal catalyst, thereby reducing cost. It also allows SOFCs to reform fuels internally, which enables the use of a variety of fuels and reduces the cost associated with adding a reformer to the system. SOFCs are also the most sulphur-resistant fuel cell type they can tolerate several orders of magnitude more sulphur than other cell types. In addition, they are not poisoned by carbon monoxide (CO), which can even be used as fuel. This allows SOFCs to use gases made from coal. 

The understanding of the interaction of S with bimetallic surfaces is a critical issue in two important areas of heterogeneous catalysis. On one hand, hydrocarbon reforming catalysts that combine noble and late-transition metals are very sensitive to sulphur poisoning [6,7]. For commercial reasons, there is a clear need to increase the lifetime of this type of catalysts. On the other hand. Mo- and W-based bimetallic catalysts are frequently used for hydrodesulphurization (HDS) processes in oil refineries [4,5,7,8]. In order to improve the quality of fuels and oil-derived feedstocks there is a general desire to enhance the activity of HDS catalysts. These facts have motivated many studies investigating the adsorption of S on well-defined bimetallic surfaces prepared by the deposition of a metal (Co, Ni, Cu, Ag, Au, Zn, A1 or Sn) onto a single-crystal face of anodier metal (Mo, Ru, Pt, W or Re) [9-29]. 

A fuel cell system would normally include a fuel desulphurising unit, using hydrodesulphurisation with ZnO, sulphm sorption (on activated carbon or a ceria-based sorbent), or a combination of both to remove sulphm from the fuel down to acceptable levels, possibly as low as a few parts per billion (ppb). Sulphur poisoning is a particular concern for fuels where desulphmisation below a few... 

Sulphur in the fuel stream, usually present as H2S, will similarly poison the anode of a PAFC. State-of-the-art PAFC stacks are able to tolerate around 50 ppm of sulphur in the fuel. Sulphur poisoning does not affect the cathode, and poisoned anodes can be reactivated by increasing the temperature or by polarisation at high potentials (i.e. operating cathode potentials). 

As already mentioned, fuel cells have a broad spectrum of fuel input sources (as shown in Fig. 9.7) the type of fuel source can be selected depending on the local conditions. With the supply infrastmcture already existent, natural gas can be the most common fuel, especially for stationary apphcations due to its easy availability, abundance, and low cost. Another benefit of using natural gas as a fuel is it can be internally reformed within the SOFC stack at temperatures as low as 600°C by steam or oxygen. However, natural gas can have sulphur-containing compounds and higher hydrocarbons which can cause sulphur poisoning and carbon deposition, respectively. 

The Ni-cermet with either stabilised zirconia, most often yttria stabilised zirconia (YSZ), or doped ceria has so far been the most successful anode in SOFCs in spite of the many problems associated with this electrode. The main reason for this is the excellent catalytic and electrocatalytic property of Ni for steam reforming of natural gas and for electrochemical oxidation of H2 and CO. These properties have been so good that they have overshadowed several drawbacks such as the sensitivity to sulphur poisoning and mechanical instability in case of redoxing. Furthermore, the electrode is fully reversible, i.e. works equally well in fuel cell and in electrolysis mode. 

The proton exchange membrane has to be improved so as to reduce resistance across the membrane and the transfer of sulphur species across the membrane that results in the deposition of sulphur on the cathode thereby poisoning the cathode. The focus will initially be on testing membranes developed for fuel cells. The NWU will be acquiring and testing fuel cell membranes which include the three top membranes in meeting the minimum requirements as set by US DOE for 2008, from Professor P. Pintauro (Vanderbilt University, Nashville, USA), Dr. R. Wycisk (Case Western Reserve University, Cleveland, USA) and Giner Electrochemical Systems Inc. 

Sulphur compounds in the fuel (0.05-0.01 %) must be removed because they would poison the reformer catalyst. Sulphur compounds are removed by using a hydrodesulphurization catalyst (HDS), typically noble metals on alumina, to convert the organic sulphur to hydrogen sulphide. Then a zinc oxide adsorbent bed is used to trap the hydrogen sulphide ... 

Commercial catalyst are available for the production of methanol and other liquid fuels from synthetic gases. The main problem is the catalyst deactivation due to chemical poisoning from chlorine and sulphur. 

Two units remove the sulphur concentrations (ppm), added to natural gas as an odorant for safety detection, or present in higher hydrocarbon feedstocks, to protect downstream catalysts (sulphur is a poison for SR catalysts) and process equipment. In particular, the organo-sulphur species are converted to H2S at pressures exceeding about 500 psig and temperatures higher than 350°C by catalytic hydrodesulphurisation (HDS unit), and Co and Mo alumina-based particulates are used as catalysts. This step is not required for methanol but would be necessary for any sulphur-containing petroleum-based fuels. A second unit permits the H2S produced in the first step to be removed by a particulate bed of ZnO. When necessary a further step for chloride removal should be included (not reported in Fig. 2.2). 

Organic sulphur- and nitrogen-compounds in motor fuels are a source for acid rain and harmful to the environment. Moreover, they are poisonous to the auto exhaust catalysts. To meet new developments in EU regulations on the S-concentration, a commonly applied one-step hydrodesulfurization (HDS), using conventional catalysts, e.g. C0-M0/7-AI2O3, is insufficient. A second HDS step, viz. a deep HDS step, can be more economical to reduce the S-content to the currently allowed European level of 350 ppm. This level will be reduced further to 50 ppm in 2005 [1]. In the first HDS step, often the heavy organic sulfur-containing polyaromatics survived, such as dibenzothiophene (DBT) and (4-, and/or 6-) alkylated DBTs [2,3]. They are the most refractory. In crude oils, there are also aromatic N-compounds, which suppress the performance of the HDS catalysts. Hence, a model feed for representative HDS-activity measurements should contain characteristic S- and N- compounds for practical relevance. 

Lead and sulphur are derived from the fuel and there is a complex equilibrium dependent upon temperatures and gas composition controlling the absorption/desorption of these poisons. In the case of lead, extended trials have demonstrated the feasibility (ref. 20) of successful operation of oxidation catalysts on leaded fuel. However, it has been noted that in the decade since introduction of lead-free fuel in the USA, residual lead levels have fallen dramatically. In that market, where leaded and unleaded fuels are both available, incidents of poisoning reflect contamination of distribution equipment or deliberate misfuelling (refs. 21,22). Sulphur may also be derived from lube oil but its impact in the sense of poisoning is low on PGM catalysts. Interaction with catalyst components can, however, influence secondary/unregulated emissions of... 

Figure 11.19 shows the process flow sheet for a pilot-scale fluidized bed gasifier, capable of processing some 20 kg/h of biomass feed, coupled with a thermal cracker and reformer reactor. The reformer is loaded with fluidizable nickel-based reforming catalyst and fitted with gas analysis ports at its inlet and outlet. The system has been used to evaluate catalyst activity and the decay of hydrocarbon conversion with time from a slip stream sample of the raw fuel gas. In this way, it is possible to quantify the frequently reported phenomenon of commercial catalyst deactivation, sometimes quite rapid, from high activity of fresh samples to lower residual activity brought about by various factors, including the presence of poisons (sulphur, chlorine) and coke formation. 

The two types of high temperature fuel cell are quite different from each other (Table 6). The molten carbonate fuel cell, which operates at 650°C, has a metal anode (nickel), a conducting oxide cathode (e.g. lithiated NiO) and a mixed Li2C03/K2C03 fused salt electrolyte. Sulphur attack of the anode, to form liquid nickel sulphide, is a severe problem and it is necessary to remove H2S from the fuel gas to <1 ppm or better. However, CO is not a poison. Other materials science problems include anode sintering and degradation, corrosion of cell components and evaporation of the electrolyte. Work continues on this fuel cell in U.S.A. and there is some optimism that the problem will be solved within 10 years. 

Key words fuel, hydrogen, methane, methanol, biogas, alkaline fuel cell (AFC), polymer electrolyte fuel cell (PEFC), phosphoric acid fuel cell (PAFC), platinum, catalyst, degradation, sulphur, carbon monoxide, poisoning, particulates. 

---