Low-temperature H 2 S removal for solid oxide fuel cell application with metal oxide adsorbents

Abstract

The desulfurization of biogas is essential for the successful operation of solid oxide fuel cells. H₂S is one of the main components in biogas. In order to feed a solid oxide fuel cell, the contaminated gas has to be reduced to a certain degree. In this work, different parameters onto the desulfurization performance of commercially available desulfurization adsorbents were investigated. The experiments were carried out using a custom made lab-scale unit. Synthetic biogas was passed through the sorbent bed and the outlet H₂S concentration was measured. Experimental runs in a fixed bed reactor were conducted to monitor H₂S removal efficiency of a zinc oxide adsorbent, an adsorbent based on a mixture of manganese and copper oxide and a zeolite adsorbent. H₂S removal efficiency was monitored under various operating conditions such as different temperatures, space velocities and inlet concentrations. This work provides useful data for adsorption tower design and process optimization.

Keywords

Desulfurization hydrogen sulfide zinc-oxide copper-oxide manganese-oxide

Introduction

This paper focuses on the low-temperature H₂S removal for solid oxide fuel cell (SOFC) application with metal oxide sorbents. SOFCs offer some important benefits and their high efficiency contributes to a diminution of CO₂ emission per unit electric power. SOFCs have the advantage of higher fuel flexibility compared to other fuel cells types (Lo Faro et al., 2012; Sasaki and Teraoka, 2003). The application of SOFCs instead of heat engines will lead to 10–20% higher efficiency in electrical power generation (Van herle et al., 2004).

Sulfur content in fuel varies from 10 ppmw in ultralow commercial gasoline to 3000 ppmw in military fuels such as JP-5 and JP-8 and 10,000 ppmw for naval distillate (Alprekin et al., 2008). Kurkela et al. (1996) expects sulphur in biogas to be predominately present as H₂S. In biogas, the content of hydrogen sulfur spans from 10–30 to 500–2000 ppm (Liu et al., 2015). Furthermore, the H₂S concentration in natural gas grids is typically limited to a few ppm to avoid facility corrosion, unnecessary production of by-products, and possible public exposure and complaints (Cosoli et al., 2008; Namgung et al., 2012). After reforming, the sulphur concentration is typically in the range of several ppmv to several hundred ppmv. Therefore, efficient sulfur removal is important for a fuel cell that uses biogas, liquefied petroleum gas or gasoline to prevent downstream sulphur poisoning of catalysts in the fuel processor and in the fuel cell anode. Even at minor quantities, the sulfur compounds can cause severe poisoning of catalysts or in fuel cell anodes (Gong et al., 2007). In order to use biogas as a fuel for a SOFC, several of its components have to be reduced to a certain degree. Among those components are, apart from H₂S, particulate matter, tar, hydrogen chloride, alkali metal species and nitrogen containing compounds such as ammonia (Aravind, 2007; Aravind et al., 2008). The influence of H₂S on the performance of SOFCs has been analysed in depth. Rasmussen and Hagen (2009) studied the influence of H₂S on the catalytic activity of a Ni/YSZ (yttrium-stabilized zirconia) electrode. After mixing the CH₄ containing fuel with H₂S in the concentration from 2 to 24 ppmv, the cell voltage decreased significantly. Matsuzaki and Yasuda (2000) reported that a H₂S concentration as low as 0.05 ppmv resulted in performance loss of a SOFC with a Ni/YSZ electrode. Veyo (1998) concluded that the impurity of 1 ppmv H₂S led to a 60 mV cell voltage drop.

There are numerous techniques available for H₂S removal from gas streams which can be classified as biological and physical–chemical processes. Biological treatments are cost effective and environmentally friendly processes. H₂S can be severed through the oxidation of microorganisms of the species thiobacillus. This oxidation requires a certain amount of oxygen which can be achieved by adding air. Furthermore, the oxidation can be carried out either in a digester or in an external scrubbing column. The biological system is able to emit large quantities of H₂S from the biogas, however, the systems’ disadvantage is its poor adaptability to fluctuating H₂S contents (Allegue and Hinge, 2014).

Physicochemical processes can be classified as reactive (chemical oxidation), absorption (caustic washing), and adsorption (metal oxides, activated carbon) techniques. The most widely used technology now is the Claus process, which recovers elemental sulfur from sulfur-containing gas. Because of thermodynamic limitations, efficient removal of H₂S from the gas is difficult and 3–5% of the H₂S is left in the tail gas (Zhang et al., 2015). Another used technology is chemical absorption. The water solubility of the H₂S is enhanced by making the water alkaline or by its oxidation to more water-soluble compounds. The crucial disadvantage of the absorption is that usually eliminates a problem with a contaminated gas stream only to create a contaminated liquid stream or a more concentrate gas liquid stream. Besides that, there are high initial investment costs as well as high consumption of chemicals. Advantages are high efficiency removal and ability to handle a wide range of pollutant concentrations. Another process based on chemical absorption is the catalytic oxidation with chelated iron salt solutions. Beside the advantage that removal efficiencies of 99.99% or higher can be reached, H₂S gets converted into elemental sulfur in this process. On the other hand, reports about plugging and foaming problems refer to a complex process control (Allegue and Hinge, 2014).

Adsorption systems are typically suitable for low gas flow rates and low H₂S concentrations. Among the available adsorbents, metal oxides are contemplated to be a valuable choice for H₂S elimination. The reaction taking place using metal oxides for sulfur removal proceeds according to the following equation:

{Me}_{x} O_{y} + y H_{2} S \to {Me}_{x} S_{y} + y H_{2} O

(1)

Desulfurization of fuel gas can be achieved by using metal oxide sorbents that form stable sulfides. Elseviers and Verelst (1999) analysed the purification performance of different metal oxides between 350 K and 923 K, and ranked Cu, Zn, Co and Mo as being the most useful. ZnO has been widely used to remove H₂S at moderate low temperatures because of its high equilibrium constant and high sulfur capacity in this temperature spectrum. Even if zinc oxide seems favourable for desulfurization among metal oxide based sorbents, its fundamental disadvantage is the limited feasibility of regeneration. For such cases, upon breakthrough, the adsorbent bed is replaced with a fresh batch of ZnO and the sulfur removal continues. CuO has an extremely high equilibrium sulfidation constant that allows an extremely low equilibrium constant even at high temperatures (Patrick et al., 1989). Due to the fact that little or no data on copper–manganese-oxide and the copper-oxide doped zeolite are present, this paper provides useful data for process optimization and adsorption tower design.

In this analysis, three commercially obtainable extrudates were evaluated for H₂S removal under various conditions. Those extrudates are ZnO, a mixture of MnO and CuO, and a CuO doped zeolite. The sulfur uptake of the different sorbents was evaluated by varying the reaction temperature and space velocity. In addition, two H₂S inlet concentrations (200 and 1000 ppmv) were used to simulate different biogas feedstocks and cover the average and maximum H₂S load of biogas from conventional biogas plants. Breakthrough behaviour of an adsorbent is also important for the adsorption process to be scaled up. As part of this research, the studies on breakthrough behaviour of the adsorbents under various operating conditions were carried out in hope that the research will lead to an optimized adsorption process being established. The goal of this study is to determine the influence of H₂S inlet concentration, reaction temperature and space velocity of three different sorbents mentioned before on sulfur uptake. The data gathered will be useful for adsorption tower design.

Experimental

Sorbents

The physical characteristics of the three different sorbents are listed in Table 1.

Figure 1.

Schematic diagram of the experimental apparatus (A: reactor; B: heating; C: valves; D: gas analyser; E: thermocouples; MFC: mass flow controller).

Table 1.

Physical characteristics of zinc oxide sorbent, manganese and copper oxide mixture sorbent, copper oxide doped zeolite.

Sorbent	ZnO	CuO–MnO	Zeolite
Bulk density (kg/m³)	1090	650	500
Pellet size (mm)	4.5	1.6	1.6
ZnO content (mass%)	87–93	0	0
CuO content (mass%)	0	>25	<25
MnO content (mass%)	0	>25	0
Aluminosilicate content (mass%)	0	0	>75

Synthetic biogas

In order to simulate different biogases, test gases with different H₂S content were used. The gases were mixed and bottled by AirLiquide. H₂S content of 200 and 1000 ppmv in nitrogen was used to simulate two different synthetic biogas feedstocks. These H₂S inlet concentrations cover the average and maximum H₂S load of biogas from conventional biogas plants. As the influence of steam content was investigated by Kim et al. (2007), only dry gas was used for the test runs. Novochinskii et al. (2004) reported that an increase of steam concentration had a negative influence on the H₂S capture capacity of the ZnO sorbent. The accuracy of the certified mixtures was 3 ppmv H₂S for the 200 ppmv gas mixture and 30 ppmv H₂S for the 1000 ppmv gas mixture.

Test rig

The custom-made lab-scale unit to perform desulfurization experiments included the following parts: a mixed and certified test gas with 200 and 1000 ppmv H₂S in nitrogen, a mass flow controller, a reactor with bypass valves, a programmable gas preheating unit and a gas analyser. A flowsheet of the test rig is represented by Figure 1. The feed gas is dosed with a mass flow controller (red-y series hi-performance, accuracy 0.3 % of full scale, +0.5% of reading). The reactor consists of two stainless steel pipes made of 1.4401 stainless steel in the following dimensions: 26.9 mm outer diameter; 21.6 mm inner diameter; each 400 mm long and connected with a union connection. The upper part of the reactor is used as a gas preheating zone. Two electric heating tapes coupled to proportional-integral derivative (PID) controllers and type K thermocouples (mounted on the outer surface of the upper and lower reactor part) ensure gas preheating. The lower reactor part contains the test cell (material: 1.4401, 21.6 mm outer diameter; 18 mm inner diameter and 150 mm length) which is suitable for taking up different sorbents. The H₂S concentration in the reactor exhaust stream was continuously monitored with an ABB AO 2000 gas analyser (accuracy 0.05 ppmv). The built-in calibration cell of the gas analyser allows calibration to be undertaken without the use of test gas bottles. SO₂ was monitored to check undesired reactions leading to SO₂ formation. The control of the mass flow controller and the two programmable heating tapes is achieved with the aid of LabVIEW software. Furthermore, the LabVIEW program was used for recording relevant process data such as temperatures and outlet H₂S concentration.

Test procedure

Different types of experiments have been performed using the ZnO sorbent, the MnO–CuO mixture sorbent and the CuO doped zeolite. At different temperatures and different Gas Hourly Space Velocities (GHSV; the quotient of the entering volumetric flow rate of the reactants divided by the catalyst bed volume), 200 and 1000 ppmv synthetic biogas was passed through the sorbent bed and the outlet H₂S concentration was measured for a 30 minutes steady state period. This investigation was carried out to find out the maximum sulfur reduction achievable from the three sorbents under different conditions. During typical test runs, sorbent pellets were loaded into the test cell and reactor, and reactor temperature was varied. The space velocity was varied through the amount of sorbent pellets, or in other words, through the height of the reactor. While the reactor was heated up to the desired temperature, the reactor was purged with nitrogen. After reaching the operating temperature, the synthetic biogas was fed into the reactor and data logging started. Before entering the gas analyser, the gas stream passed a sample gas cooler consisting of a temperature controller, a heat exchanger and a peristaltic pump.

Due to the fact that except for the ZnO sorbent few or no data is available, experiments to characterize the maximum sulfur capture capacity had to be carried out. Since sulphur saturation can take an extremely long time, 1000 ppmv H₂S in nitrogen was used to characterize the sulfur capacity of the sorbents. For those experiments, 1 g of sorbent sample has been flowed through at 293 K and a space velocity of 4000 h⁻¹ for 24 hours. The sulfur capture capacity of the sorbent samples was calculated using the following formula, which has been used by Novochinskii et al. (2004) and Kim et al. (2007). In addition, the results were cross checked by analytical chemistry.

S_{cap} = \frac{FR \cdot \int_{0}^{t} (C_{H_{2} S_{in}} - C_{H_{2} S_{out}}) d_{t} \cdot 32 \cdot 10^{- 4}}{V_{mol} \cdot m_{sample}}

(2)

S_cap describes the sulfur capture capacity in (g of S)/(100 g of sorbent). The other symbols shown in equation (2) are explained in the nomenclature.

Nomenclature
Symbol	Units	Description
FR	L/h	gas flow rate
T	h	experiment time
V _mol	L/mol	gas molar volume
$C_{H_{2} S_{in}}$	ppmv	H₂S inlet concentration
$C_{H_{2} S_{out}}$	ppmv	H₂S outlet concentration
m_s_ample	g	sorbent sample weight
32	g/mol	molar weight of sulfur

Results and discussion

Effect of the reaction temperature

Experiments were conducted at different temperatures to determine the optimum temperature providing the highest adsorption performance. Inlet H₂S concentration as low as 200 ppmv was selected to simulate a biogas feedstock with average sulfur content. The technical data for the used pellets is given in Table 1. The results of adsorption experiments showing the effect of temperature are shown in Figures 2 to 4. Figure 2 shows the effect of reaction temperature on H₂S outlet concentration of ZnO adsorbent. Figure 3 shows the effect of reaction temperature on H₂S outlet concentration of CuO–MnO adsorbent and Figure 4 shows the results for zeolite. The H₂S outlet concentration for all sorbents decreases when the temperature is increased. The sulfur removal efficiency of the sorbent used is defined as the amount of H₂S adsorbed at the sorbents surface related to the H₂S concentration at the outlet. An increase of the reaction temperature from 293 K to 473 K led to a decreased H₂S concentration at the reactor outlet. In numbers, the efficiency increased from 91.93% to 98.72% for ZnO, from 94.26% to 98.71% for the CuO–MnO sorbent and from 98.28% to 99.42% for zeolite.

Figure 2.

Effect of reaction temperature on H₂S outlet concentration of ZnO adsorbent. Space velocity 4000 h⁻¹; H₂S inlet concentration 200 ppmv.

Figure 3.

Effect of reaction temperature on H₂S outlet concentration of CuO–MnO adsorbent. Space velocity 4000 h⁻¹; H₂S inlet concentration 200 ppmv.

Figure 4.

Effect of reaction temperature on H₂S outlet concentration of zeolite adsorbent. Space velocity 4000 h⁻¹; H₂S inlet concentration 200 ppmv.

The sulfur removal efficiency for ZnO extrudates observed in this study is in line with the literature (Kim et al., 2007). The fact that the sulfur removal efficiency is not as high as in the cited literature might be to the result of the following reason: The reaction of ZnO with H₂S is favoured kinetically as the temperature increases. Kim et al. (2007) achieved a sulphur removal efficiency of 99.99% at 636 K. Due to the fact that there was no data available, the sorbent based on CuO and MnO in combination with zeolite could not be compared with literature.

Effect of space velocity

Figure 5 shows the effect of the space velocity upon H₂S breakthrough on ZnO extrudates. All test runs were carried out with synthetic biogas with 200 ppmv H₂S concentration and at a reaction temperature of 293 K. Increasing space velocity from 4000 h⁻¹ to 8000 h⁻¹ elevated the H₂S outlet concentration dramatically from 20 ppmv to 70 ppmv after the monitored 30 minutes. Increasing the space velocity from 8000 h⁻¹ to 12,000 h⁻¹ had little impact and only slightly increased the H₂S outlet concentration from 70 ppmv to 80 ppmv after the monitored time span of 30 minutes. These results propose that S_cap of zinc oxide is restricted by the contact between gas and sorbent in the fixed bed reactor at higher space velocities.

Figure 5.

Effect of space velocity on H₂S outlet concentration of ZnO adsorbent. Reaction temperature 293 K; H₂S inlet concentration 200 ppmv.

Figure 6 exhibits the obtained test results for the CuO–MnO sorbent. It was found that the impact of the space velocity is minor, compared to the experimental runs with ZnO extrudates. Doubling the space velocity from 4000 h⁻¹ to 8000 h⁻¹ elevated the H₂S outlet concentration from 12.5 ppmv to 13 ppmv after 30 minutes. Increasing the space velocity up to 12,000 h⁻¹ only slightly increased the outlet concentration by about 1 ppmv. Comparing the curves in Figure 6, indicates that the gas solid contact has no limiting effect at the tested space velocities.

Figure 6.

Effect of space velocity on H₂S outlet concentration of CuO–MnO adsorbent. Reaction temperature 293 K; H₂S inlet concentration 200 ppmv.

The results for the zeolite tested under various space velocities can be found in Figure 7. Comparing to the other sorbents, the zeolite achieved the best H₂S outlet concentrations under all monitored space velocities. Doubling the space velocity from 4000 h⁻¹ to 8000 h⁻¹ increased the outlet concentration from 3 to 6 ppmv. The sulfur removal efficiency of tested sorbents at different space velocities changes as follows: An increase of the space velocity from 4000 h⁻¹ to 12,000 h⁻¹ led to a decreased sulfur removal efficiency from 91.93% to 70.77% for ZnO. Increasing the space velocity from 4000 h⁻¹ to 12,000 h⁻¹ for the CuO–MnO adsorbent slightly decreased the sulphur removal efficiency from 94.26% to 93.28%. The zeolite sulfur removal efficiency dropped from 98.28% to 95.58%. It can be seen that the space velocity has a greater influence on the ZnO sorbent than on the CuO–MnO sorbent and the zeolite.

Figure 7.

Effect of space velocity on H₂S outlet concentration of zeolite adsorbent. Reaction temperature 293 K; H₂S inlet concentration 200 ppmv.

Effect of inlet H₂S concentration

Figures 8 to 10 show the effect on H₂S outlet concentration for synthetic biogases with 200 and 1000 ppmv H₂S inlet concentration. Increasing the inlet concentration from 200 2to 1000 ppmv on the ZnO extrudates had major impact on the H₂S outlet concentration. In numbers, the H₂S outlet concentration increased from 18 ppmv to 270 ppmv after 30 minutes. On the other hand, Figure 9 shows the impact of increasing the inlet concentration on the CuO–MnO extrudate. The increase of the H₂S inlet concentration resulted in an increase of the outlet concentration from 12 ppmv to 68 ppmv. As the zeolite is concerned, the H₂S outlet concentration increased from 4 ppmv to 51 ppmv. This can be seen in Figure 10. Comparing the outlet concentrations of the experiments indicates that the zeolite works best at high inlet concentrations. It can be seen that CuO–MnO reaches a steady state after 5–6 minutes while ZnO and the zeolite still show an increase after 30 minutes. In numbers, the sulfur removal efficiency for ZnO decreases from 91.93% at 200 ppmv H₂S inlet concentration to 82.46% at 1000 ppmv H₂S inlet concentration. For CuO–MnO sorbent it decreases from 94.26% to 93.52% and for the zeolite it decreases from 98.28% to 96.83%.

Figure 8.

Effect of H₂S inlet concentration on ZnO adsorbent on H₂S outlet concentration. Reaction temperature 293 K; space velocity 4000 h⁻¹.

Figure 9.

Effect of H₂S inlet concentration on CuO–MnO adsorbent on H₂S outlet concentration. Reaction temperature 293 K; space velocity 4000 h⁻¹.

Figure 10.

Effect of H₂S inlet concentration on zeolite adsorbent on H₂S outlet concentration. Reaction temperature 293 K; space velocity 4000 h⁻¹.

Maximum sulfur capacity

All the tests carried out under different operating conditions for 30 minutes give an idea how the modification of one parameter affects the H₂S outlet concentration. Long time experiments at 293 K and a space velocity of 4000 h⁻¹ were performed to characterize the maximum sulfur capture capacity of the sorbents. The results can be seen in Figure 11 and Table 2. S_cap of the sorbents was calculated by equation (2) and cross checked by analytical chemistry. The CuO–MnO sorbent showed the best sulfur capture capacity with 180.1 (mg sulfur/g adsorbent). The zeolite achieved 31 and the ZnO was measured and calculated at 46.3 (mg sulfur/g adsorbent).

Figure 11.

Sulfur capture capacity. Reaction temperature 293 K; space velocity 4000 h⁻¹; H₂S inlet concentration 1000 ppmv.

Table 2.

Sulfur capture capacity in (mg of sulfur)/(g sorbent). Reaction temperature 293 K; space velocity 4000 h⁻¹; H₂S inlet concentration 1000 ppmv.

Sorbent	ZnO	CuO–MnO	Zeolite
Sulfur capture capacity (mg/g)	46.3	180.1	31

Conclusion

The desulfurization of biogas is essential for the successful operation of SOFCs. H₂S adsorption on three different extrudates, one based on ZnO, one based on a mixture of MnO and CuO, and a copper doped zeolite were evaluated at low temperatures in a fixed bed reactor under various conditions. The sulfur capture capacity S_cap, a value for the extent of sorbent utilization for sulfur removal, was evaluated. This experimental study on low-temperature H₂S adsorption, using different sorbents, has led to the following conclusions:

With the inlet temperature increasing from 293 to 373 to 473 K, S_cap increased for all sorbents. The increase of the temperature slightly improved the performance of CuO–MnO and the zeolite, and had a major impact on ZnO. The authors are aware of the fact that ZnO works best at elevated temperatures, but decided to use it because it is well known and studied and may provide a comparison to the other tested sorbents. The efficiency of sulfur removal is kinetically favoured at higher temperatures, while its equilibrium is restricted to lower temperatures. The operating temperature influences both the diffusion rate and the thermodynamic equilibrium. This is consistent with the reports in the literature.

Decreasing the space velocity from 12,000 to 8000 to 4000 h⁻¹ led to the following results for the ZnO sorbent. At the space velocities 8000 h⁻¹ and 12,000 h⁻¹, the H₂S outlet concentration dramatically increased compared to the test runs obtained at 4000 h⁻¹. The test at a space velocity of 4000 h⁻¹ resulted in the lowest H₂S outlet concentrations due to the increased contact time between gas and sorbent inside the reactor. The sorbent based on a mixture of CuO and MnO as well as the zeolite were monitored the same way. The test runs at 12,000 h⁻¹, 8000 h⁻¹ and 4000 h⁻¹ resulted in only a small difference in the H₂S outlet concentration.

The inlet H₂S concentration was increased from 200 to 1000 ppmv to simulate two different biogas feedstocks. All tested sorbents showed an increased sulfur capture rate. This demonstrates that the extent of sorbent utilization was improved by the sulphur content of the synthetic biogas. Increasing the H₂S concentration from 200 to 1000 ppmv had major impact on the H₂S outlet concentration of the ZnO extrudates compared to the CuO–MnO sorbents and the zeolite.

Even if the zeolite shows very good results at the tested temperatures, space velocities and inlet concentrations its maximum sulfur capacity is very low. This results in big adsorbers compared to CuO–MnO sorbent.

Of all the experiments conducted the conclusion is that operating the zinc oxide sorbent around a temperature of 473 K and a space velocity of 4000 h⁻¹ leads to the best results for the H₂S removal from synthetic biogas. For the sorbent based on a mixture of manganese and copper oxide, the working temperature should be elevated. The space velocity has no major impact. The zeolite achieved the best H₂S outlet concentrations and the lowest sulfur capture capacity. Direct comparison between the monitored sorbents shows that CuO–MnO has the highest sulfur capture capacity and should be considered as a desulfurization material at lower temperatures. In a future step, the interference of hydrogen chloride on the sorbents will be investigated.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to thank the Austrian Research Promotion Agency (FFG) for funding this project. The project title is SOFCool (grant number 843835).

References

Allegue LB and Hinge J (2014) Biogas upgrading: Evaluation of methods for H₂S removal. Technical report, Danish Technological Institute.

Alprekin G, Jayaraman A, Dubovik M, et al. (2008) Desulfurization of logistic fuels for fuel cell APUs. Technical report.

Aravind PV (2007) Studies on high efficiency energy systems based on biomass gasifiers and solid oxide fuel cells with Ni/GDC anodes. PhD Thesis, TU Delft, Delft.

Aravind

Ouweltjes

Woudstra

(2008) Impact of biomass-derived contaminants on SOFCs with Ni/gadolinia-doped ceria anodes. Electrochemical and Solid-State Letters 11(2): B24, DOI: 10.1149/1.2820452.

Cosoli

Ferrone

Pricl

(2008) Hydrogen sulphide removal from biogas by zeolite adsorption: Part I. GCMC molecular simulations. Chemical Engineering Journal 145(1): 86–92. DOI: 10.1016/j.cej.2008.07.034.

Elseviers

Verelst

(1999) Transition metal oxides for hot gas desulphurisation. Fuel 78(5): 601–612. DOI: 10.1016/S0016-2361(98)00185-9.

Gong

Liu

Trembly

(2007) Sulfurtolerant anode materials for solid oxide fuel cell application. Journal of Power Sources 168(2): 289–298. DOI: 10.1016/j.jpowsour.2007.03.026.

Kim

Jeon

(2007) Removal of hydrogen sulfide from a steam-hydrogasifier product gas by zinc oxide sorbent. Industrial & Engineering Chemistry Research 46(18): 5848–5854. DOI: 10.1021/ie0614531.

Kurkela E, Laatikainen J, Sthlberg P, et al. (1996) Pressurized Fluidized-Bed Gasification Experiments With Wood, Peat and Coal at VTT in 1991–1992; Pt. 3. Espoo: VTT. ISBN 951-38-4949-X 978-951-38-4949-8.

10.

Liu

Zhang

Wei

(2015) Selective removal of H₂S from biogas using a regenerable hybrid TiO₂/zeolite composite. Fuel 157: 183–190. DOI: 10.1016/j.fuel.2015.05.003.

11.

Lo Faro

Antonucci

(2012) Fuel flexibility: A key challenge for SOFC technology. Fuel 102: 554–559. DOI: 10.1016/j.fuel.2012.07.031.

12.

Matsuzaki

Yasuda

(2000) The poisoning effect of sulfur-containing impurity gas on a SOFC anode: Part I. Dependence on temperature, time, and impurity concentration. Solid State Ionics 132(34): 261–269. DOI: 10.1016/S0167- 2738(00)00653-6.

13.

Namgung

Ahn

Song

(2012) Development of a two-phase bioreactor for the biological removal of hydrogen sulfide from biogas. Energy Procedia 14: 1143–1148. DOI: 10.1016/j.egypro.2011.12.1067.

14.

Novochinskii

Song

(2004) Low-temperature H₂S removal from steam-containing gas mixtures with ZnO for fuel cell application. 1. ZnO particles and extrudates. Energy & Fuels 18(2): 576–583. DOI: 10.1021/ef030137l.

15.

Patrick

Gavalas

Flytzani-Stephanopoulos

(1989) High-temperature sulfidation-regeneration of copper(II) oxide-alumina sorbents. Industrial & Engineering Chemistry Research 28(7): 931–940. DOI: 10.1021/ie00091a008.

16.

Rasmussen

JFB

Hagen

(2009) The effect of H₂S on the performance of NiYSZ anodes in solid oxide fuel cells. Journal of Power Sources 191(2): 534–541. DOI: 10.1016/j.jpowsour.2009.02.001.

17.

Sasaki

Teraoka

(2003) Equilibria in fuel cell gases. II. The C–H–O ternary diagrams. Journal of the Electrochemical Society 150(7): A885–A888. DOI: 10.1149/1.1577338.

18.

Van herle

Marchal

Leuenberger

(2004) Process flow model of solid oxide fuel cell system supplied with sewage biogas. Journal of Power Sources 131(12): 127–141. DOI: 10.1016/j.jpowsour.2004.01.013.

19.

Veyo SE (1998) Evaluation of fuel impurity effects on solid oxide fuel cell performance. Technical report, Westinghouse Electric Company.

20.