Oxygen removal from syngas by catalytic oxidation ... - Semantic Scholar

Journal of the Energy Institute 87 (2014) 246–252

Contents lists available at ScienceDirect

Journal of the Energy Institute journal homepage: http://www.journals.elsevier.com/journal-of-the-energyinstitute

Oxygen removal from syngas by catalytic oxidation of copper catalyst Jun Han a, b, Xiang He a, Rui Li b, Caixia Wan c, Qiangu Yan b, Fei Yu b, * a

Hubei Key Laboratory of Coal Conversion and New Material, Wuhan University of Science and Technology, Wuhan 430081, China Department of Agriculture and Biological Engineering, Mississippi State University, Starkville, MS 39762, USA c Department of Bioengineering, University of Missouri, Columbia, MO 65211, USA b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 16 August 2013 Accepted 29 January 2014 Available online 2 April 2014

The syngas from biomass gasification may contain trace oxygen besides of gaseous hydrocarbons, which will result in the temporarily or even permanently deactivation of Fischer–Tropsch (F–T) catalysts. In this paper, CuO–CeO2/Al2O3 catalyst was developed to efficiently remove the trace oxygen from biomass syngas. The experimental results demonstrated that CuO–CeO2/Al2O3 catalyst was considerably effective in removing oxygen to the level of below 1 ppm, its lifetime and deoxygenation capacity were 160 h and 3000 ml/g, respectively. Moreover, the optimum conditions of CuO–CeO2/Al2O3 catalyst were 200 C, 3.45 105 Pa, and 3000 h1 gas hourly space velocity. Ó 2014 Energy Institute. Published by Elsevier Ltd. All rights reserved.

Keywords: Syngas Oxygen Fischer–Tropsch Removal Deactivation catalyst

1. Introduction The synthesis of transportation fuel has been the hot topic of biomass utilization due to the depletion of oil, national energy security and environmental issues. At present, the most potential routes for producing transportation fuels from biomass is gasification, followed by Fischer–Tropsch (FT) synthesis [1]. However, the syngas from biomass gasification may contain some trace pollutants such as 10,000– 15,000 ppm tars, 2000–4000 ppm ammonia, 100–500 ppm H2S [2,3] and 0.1–1% oxygen [4,5] besides gaseous hydrocarbons. Unfortunately, it has been discovered that trace oxygen can temporarily or even permanently deactivate the commercial FT catalysts. Yan et al. reported that the lifetime of F–T synthesis catalyst was only 40 h without the cleaning process. However, the lifetime of catalyst would be over 1000 h after the oxygen in the syngas was removed to the level below 1 ppm, as shown in Fig. 1 [6]. It was reported that the catalyst oxidization by precious metals could remove oxygen to a level of below 0.02 ppm due to the reaction between oxygen and hydrogen or carbon monoxide. The mechanism of the H2 oxidation reaction over Pt surface was studied by using mass spectrometry and laser-induced fluorescence [7]. Lei et al. pointed out that oxygen content in hydrogen could be reduced from 0.4% to 2 106–3 106 (v/v) in the reaction temperature range of 150–170 C with a space velocity of 5000 h1. Meanwhile, the increase reaction temperature had little influence on the removal efficiency of oxygen [8]. However, the widely application of removing oxygen from the syngas was suppressed because of the high cost of noble metals and sulfur poisoning. CuO is a cheap catalyst with high activity, which has been also widely used in a variety of catalytic reactions including the oxidation of carbon monoxide and hydrocarbons [9]. Avgourpoulos et al. reported that CuO–CeO2 catalyst had the same activity as Pt/Al2O3 catalyst [10]. Wan et al. pointed out that Cuþ species were active phase after the pretreatment by CO [11]. In the presence of rich hydrogen, hydrogen molecules were firstly absorbed on the catalyst surface, followed by disassociating from the catalyst surface and forming hydrogen atoms. At the same time, oxygen also reacted with catalyst and formed the oxides of O2. Lastly, hydrogen atoms easily reacted with O2 and formed water [11]. In the system of rich CO, the reaction complies with the Langmuir–Hinshelwood mechanism [12]. During catalytic processes, the adsorbed oxygen reacts with carbon monoxide, which results in forming a CO3 complex as an intermediate stage. The reaction releases much heat and leads to the sharply increase of temperature. The high temperature may change the crystal and pore structure of the deoxidant catalyst, which may lead lower the dispersion caused by the sintering. Zhu [13] and Zeng [14] stated that Au/CeO2 and CuO/CeO2 catalyst had a strongly active for CO oxidation due to their oxygen storage capacity. BASF also proved that CuO catalyst tablets could remove oxygen or CO/H2 from gas stream [15]. * Corresponding author. E-mail address: [email protected] (F. Yu). http://dx.doi.org/10.1016/j.joei.2014.03.005 1743-9671/Ó 2014 Energy Institute. Published by Elsevier Ltd. All rights reserved.

J. Han et al. / Journal of the Energy Institute 87 (2014) 246–252

247

Fig. 1. Comparison of the catalyst lifetime with and without clean process.

Table 1 The compositions of catalyst. Name

Ce (wt %)

Cu (wt %)

Al2O3 (wt %)

Catalyst F Catalyst P

5 0

10 10

85 90

Ge prepared Mn–Cu–Co/montmorillonite deoxidant by co-precipitation, which could reduce oxygen content from 500 106 to 0.01 106 (v/v) at a space velocity of 1000–8000 h1, and the operation temperature range of 25–200 C [16]. Ren et al. also developed a Mn–Cu–Co/active carbon deoxidant which had a high selectivity of conversing CO to CO2. After the clean process, the oxygen concentration was below 1 ppm [17]. In this paper, copper catalysts with CeO2 promoter and alumina support were developed. The optimum operations and their catalyst performances were also investigated. 2. Experimental 2.1. Material and method It was discovered that the oxides of transition metals combined with other oxides such as cerium had the active radicals – hydroxyl roots, which could store oxygen in its structure. Moreover, Ceria could promote the dispersion of copper on alumina support [18]. Therefore, the activity of catalyst was increased when CeO2 was incorporated into CuO lattice. In this experiment, CeO2 and alumina were used as promoter and support, respectively. According to the experimental results of Strohmeier [19], the highly dispersed Cu oxide species were well interacted with support alumina when Cu content in the catalyst was 10%wt. Three catalysts with different components were prepared, and the compositions are listed in Table 1. The catalysts were prepared by a two-step impregnation method. As for catalyst F, the required amount of cerium nitrate [Ce(NO3)3$6H2O] (99.5%, Acros Organics) solution was firstly impregnated over the g-alumina (surface area, 246 m2/g, pore volume, 1.15 mL/g), then the precursor was dried at 120 C for 6 h, followed by calcinated at 600 C for 6 h. Thus, CeO2/Al2O3 was prepared. During the subsequent impregnation, the required amount of copper nitrate [Cu(NO3)2$3H2O] (99.7%, Sigma–Aldrich) solution was impregnated over the prepared CeO2/Al2O3 and dried under 120 C for 6 h, followed by calcinations under 350 C for 6 h. The final catalyst was prepared. The properties of the catalysts were listed in Table 2. 2.2. Pretreatment The reduction can activate surface copper oxide into metal copper, which increases the catalyst activity and its oxygen capacity. Wan et al. [11] demonstrated that the deoxygenation efficiency of the CO-pretreated catalyst was higher than that of fresh catalysts. In these runs, catalyst F was active by CO–N2 stream (20% CO by volume, N2 was balanced gas) at 250 C for 6 h under atmosphere pressure. During the reduced process, the gas flow rate was kept about 20 ml/min. After the reduction, the reactor was purged with N2 stream until the treated catalyst was cooled to a determined temperature. Lastly, the catalyst was directly exposed to the syngas when the reactor temperature was stabilized. Table 2 The properties of the catalyst. Name

BET surface area (m2/g)

Pore volume (cm3/g)

Average pore diameter (mm)

Catalyst F Catalyst P

129 187

0.31 0.64

4.7 7.5

248


Fig. 2. Schematic diagram of experimental apparatus.

2.3. Catalytic activity Catalytic activity tests were performed in a silica tubular reactor with an inside diameter of 15.24 cm and a length of 50 cm. The reactor was placed in a tube furnace with electric heating, and the temperature was controlled by thermocouples, as shown in Fig. 2. Before the test, 3 g catalyst was placed at the center of reactor. The gases were supplied by cylinder, which came from a wood gasifier in Mississippi State University. The flow rates of the reaction gases and reduction gases were controlled by the mass flow controllers. During the tests, the gas hourly space velocity (GHSV) was set from 500 h1 to 5000 h1. The reaction temperature was determined at 100 C, 150 C, 200 C and 250 C under the pressure of 0 Pa and 3.45 105 Pa. The oxygen conversion is defined as Eq. (1).

CO2 ¼

Xin Xout Xout 100% ¼ 1 Xin Xin

(1)

where C is the conversion rate (%) and X is the (inlet or outlet) oxygen concentration (ppm). In Eq. (2), the deoxygenation capacity (DC) is the total amount of syngas cleaned that the residual oxygen concentration is below 1 106 (v/v) every g catalyst.

Deoxygenation capacityðml=gÞ ¼

Total exhast oxygen at outlet below 1 ppmðmlÞ Weight of catalystðgÞ

Fig. 3. TG/DTG of (a) and (b) as a function of temperature. (a) P and (b) F.

(2)


249

3. Results and discussion 3.1. Thermogravimetric analysis Thermogravimetric analysis was conducted in a Shimadzu TGA-50H to determine the calcination temperature. The weight variation of catalyst was recorded as a function of temperature up to 1000 C. In the experiment, the precursor of catalyst before calcination (after drying at 120 C) was analyzed. The heating rate and nitrogen flow rate were 10 C/min and 50 ml/min, respectively. Fig. 3a showed that the weight loss of catalyst P from 100 C to 200 C could be attributed to the release of physically adsorbed water. The decomposition of nitrite and the release of NO2 and O2 Eq. ((3)) [20] was accounted for the remarkable weight loss from 300 C to 600 C. When the temperature was above 600 C, no obviously weight loss can be observed. On the basis of above data, it was concluded that the optimum calcination temperature of catalyst P should be above 600 C. As for catalyst F, the weight loss can be ignored under 100 C, which suggested that the precursor had been fully dried at 120 C. The crystallization water was released at the temperature rage of 200–250 C and the nitrates decomposition took place between 250 C and 350 C (Eq. (4)) [21]. After 350 C, TG curve was flat. Therefore, it can be concluded that the calcination temperature of catalyst F shouldn’t be lower than 350 C.

2CuðNO3 Þ2 /2CuO þ 4NO2 þ O2

(3)

2CeðNO3 Þ3 /2CeO2 þ 6NO2 þ O2

(4)

3.2. Effect of temperature and pressure on the catalyst performance A 3 g of catalyst F and catalyst P were loaded in the silica reactor, respectively. In these runs, the reaction temperatures range was 100– 250 C. The operating pressure was atmosphere pressure and 3.45 105 Pa. Moreover, the gas hourly space velocity (GHSV) was maintained at 3000 h1. The syngas used in the runs contained 46% N2, 21% CO, 18% H2, 12% CO2, 2% CH4 and 0.9% O2, which came from a wood gasifier in Mississippi State University. Fig. 4 presented that both catalyst P and catalyst F had good deoxygenation efficiency. When the temperature was above 200 C, the concentration of oxygen at the outlet could be ignored. Hence, the oxygen conversions were almost 100%. At the same time, it was found that the increase of temperature and pressure could promote the removal efficiency of oxygen. At an atmosphere pressure, the oxygen concentration at the outlet of catalyst P was 340 ppm under 100 C, which was decreased to 0 ppm when the temperature was increased to 200 C. In order to study the effects of pressure, the oxygen removal experiments were carried out at 100 C under 0 and 3.45105 Pa, and it was found the oxygen concentrations under 0 and 3.45105 Pa were 340 and 7 ppm, respectively. Therefore, the optimum operation conditions of catalyst P were 150 C and 3.45 105 Pa. As for catalyst F, the oxygen concentration at the outlet was below 10 ppm under 100 C and 3.45 105 Pa. The increase of pressure shifted the equilibrium towards the shrinking volume side of the reaction, which favored the formation of water. Fig. 4 also demonstrated that catalyst F had a higher catalytic activity than that of catalyst P. The reasons maybe that there is an interaction between copper and cerium oxide, which promotes the dispersion of CuO, the formation and stability of abundant Cuþ species, the production of oxygen vacancies, and lattice oxygen of copper oxide [10]. 3.3. Variation of syngas composition after removing oxygen

C þ O2 /CO2

ðDH ¼ 405:9 kJ=molÞ

(5)

Fig. 4. Effect of pressure and temperature on F and P.

250


Fig. 5. The variation of syngas composition after removing oxygen.

1 H2 þ O2 /H 2 O 2

ðDH ¼ 242 kJ=molÞ

(6)

According to Eqs. (5) and (6), the maximum consumption of H2 was about 1.8% if oxygen was completely removed (the oxygen concentration in syngas was 0.9%). However, Fig. 5a presented that the variation of H2 concentration was only about 1%. Hence, it can be concluded that there was limited CO2 formation during the oxygen reduction, which was proven by Fig. 5b. Park et al. stated that the CO

Fig. 6. Effect of gas hourly space velocity (GSHV) on deoxygenation efficiency at 150, 180 and 200 C, 0 Pa over catalyst F.


251

Fig. 7. Effect of temperature on F at 170 C and 200 C, 3.45 105 Pa, and gas hourly space velocity is 3500 h1.

oxidation activity decreased with increasing CO2 content due to the competitive adsorption of CO and CO2 on the ceria surface and carbonate formation, which prevented the participation of oxygen [22]. In order to decrease copper carbonate, they declared that the reaction temperature should be above 180 C. Fig. 5 also presented that the maximum consumption of H2 occurred at 3.45 105 Pa and 100 C when catalyst F was used as catalyst. Moreover, it was also found that the increase of pressure could promote the consumption of H2 during the oxygen catalytic removal. However, the consumption of CO seemed to be independent of the reaction temperature. At the same time, the increase of reaction pressure had a minor negative influence on CO oxidation reaction.

3.4. Effect of space velocity on the catalyst performance Fig. 6 displayed that the high space velocity had a detrimental effect on the oxygen removal. In this run, the initial oxygen concentration was 0.9%, and the gas hourly space velocity differed from 1500 to 4000 h1. The reaction temperatures were 150, 180 and 200 C, respectively. The operation pressure was 0 Pa. The experimental results discovered that the high space velocity favored the diffusion of oxygen, which increased the reaction rate and had a positive influence on the oxygen removal. However, the space velocity was too high, which would result in oxygen penetration. Consequently, the residual oxygen concentration was increased sharply.

3.5. Deoxygenation capacity In this run, the reaction temperatures were 170 and 200 C, respectively. The operating pressure was 3.45 105 Pa and the gas hourly space velocity was maintained at 3500 h1. Before the experiment, 9 g catalyst F was loaded into the reactor. At the same time, the syngas came from a wood gasifier was used as reaction gas, which contained 46% N2, 21% CO, 18% H2, 12% CO2, 2% CH4 and 0.9% O2. Fig. 7 showed that the lifetime of catalyst F under 170 C and 200 C were 40 h and 160 h, respectively. At the same time, the deoxygenation capacity under 170 C and 200 C were 677 ml/g and 2890 ml/g (the deoxygenation capacity was defined in Eq. (2)). Although copper catalyst lifetime was shorter than that of the noble metals, copper catalyst was cheaper. Fig. 7 also presented that catalyst F had a longer life under 200 C than that under 170 C. There are two mechanisms accounted for the above phenomena: one is that the deactivated catalysts can be regenerated by syngas at 200 C; Another reason is that the accumulation of carbonates on catalyst would be decomposed by calcination and the copper oxide species could be decreased at 200 C.

4. Conclusions The syngas from biomass gasification may contain some trace pollutants such as 10,000–15,000 ppm tars, 2000–4000 ppm ammonia, 100–500 ppm H2S and 0.1–1% oxygen, besides of gaseous hydrocarbons. Unfortunately, it has been discovered that trace oxygen can temporarily or even permanently deactivate the commercially FT catalysts. In order to remove oxygen from syngas to 1 ppm, copper catalyst was developed in this paper. The experimental results discovers that copper is an effective catalyst for removing oxygen from syngas, and its lifetime and deoxygenation capacity are 160 h and 3000 ml/g, respectively. The optimum conditions of copper catalyst are 200 C, 3.45 105 Pa, and 3000 h1 gas hourly space velocity.

Acknowledgement Funding for this work was provided by the U.S. Department of Energy under Grant DOEDE-FG36-06GO86025 and DE-FC2608NT01923, and the U.S. Department of Agriculture under Award (AB567370MSU).

252


References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]

R. Luque, A.R. de la Osa, J.M. Campelo, A.A. Romero, J.L. Valverde, P. Sanchez, Energy Environ. Sci. 5 (2012) 5186–5202. H. Leibold, A. Hornung, H. Seifert, Powder Technol. 180 (2008) 265–270. Q. Yan, C. Wan, J. Street, D.W. Yan, J. Han, F. Yu, Bioresour. Technol. 147 (2013) 117–123. G. Pu, H.-P. Zhou, G.-T. Hao, Int. J. Hydrogen Energy 38 (2013) 15757–15763. M.S. Saha, Y. Zhang, M. Cai, X. Sun, Int. J. Hydrogen Energy 37 (2012) 4633–4638. Q. Yan, C. Wan, J. Liu, J. Gao, F. Yu, J. Zhang, Z. Cai, Green. Chem. 15 (2013) 1631–1640. W.R. Williams, C.M. Marks, L.D. Schmidt, J. Phys. Chem. 96 (1992) 5922–5931. J.-M. Lei, H.-z. Ling, B. Chen, Z. Liao, Nat. Gas. Chem. Ind. (2005) 30. D. Terribile, A. Trovarelli, C. de Leitenburg, A. Primavera, G. Dolcetti, Catal. Today 47 (1999) 133–140. G. Avgouropoulos, T. Ioannides, C. Papadopoulou, J. Batista, S. Hocevar, H.K. Matralis, Catal. Today 75 (2002) 157–167. H. Wan, Z. Wang, J. Zhu, X. Li, B. Liu, F. Gao, L. Dong, Y. Chen, Appl. Catal. B: Environ. 79 (2008) 254–261. C. Laberty, C. Marquez-Alvarez, C. Drouet, P. Alphonse, C. Mirodatos, J. Catal. 198 (2001) 266–276. P. Zhu, M. Liu, R. Zhou, India J. Chem. 51A (2012) 1529–1537. S. Zeng, Y. Wang, K. Liu, F. Liu, H. Su, Int. J. Hydrogen Energy 37 (2012) 11640–11649. BASF, Adsorbents for Oxygen Removal. http://www.adsorbents.pro/application/oxygen_removal.aspx. Q. Ge, Master, Nanjing University of Science and Technology, 2006. J. Ren, Y. Li, X. Sun, Chem. Catal. Methanol Technol. (2000) 56–58. A. Alouche, Jordan J. Mech. Industr. Eng. 2 (2008) 111–116. B.R. Strohmeier, D.E. Levden, R.S. Field, D.M. Hercules, J. Catal. 94 (1985) 514–530. R.I. Olivares, W. Edwards, Thermochim. Acta 560 (2013) 34–42. J.G. Jackson, R.W. Fonseca, J.A. Holcombe, Spectrochim. Acta Part B At. Spectrosc. 50 (1995) 1449–1457. J.W. Park, J.H. Jeong, W.L. Yoon, Y.W. Rhee, J. Power Sources 132 (2004) 18–28.