Catalysts and catalytic processes for the production of hydrogen-rich gas for fuel cell feeding Текст научной статьи по специальности «Химические науки»

CHEMICAL PROBLEMS 2019 no. 2 (17) ISSN 2221-8688

193

UDC 544.478-03

CATALYSTS AND CATALYTIC PROCESSES FOR THE PRODUCTION OF HYDROGEN-RICH GAS FOR FUEL CELL FEEDING

S.D. Badmaev1,2, V.D. Belyaev1,2, M.V. Konishcheva1,2, A.V. Kulikov1,3, A.A. Pechenkin1,2, D.I. Potemkin1,2, V.N. Rogozhnikov1, P.V. Snytnikov1,2,3, V.A. Sobyanin1

1Boreskov Institute of Catalysis, 630090, Novosibirsk, Pr. Akademika Lavrentieva, 5 2 Novosibirsk State University, 630090, Novosibirsk, St.Pirogova, 2 3 «UNICAT» LLC, 630090, Novosibirsk, Pr. Akademika Lavrentieva, 5 e-mail: sobyanin@catalysis.ru

Received 28.05.2019

Abstract: The paper considers new approaches to the preparation of structured catalysts and peculiarities of the processes for the production of hydrogen-rich gas from hydrocarbons and dimethyl ether for fuel cell feeding applications. The results of the studies of structured catalyst for CO removal from reformate by selective methanation are presented.

Keywords: partial oxidation, dimethyl ether, structured catalyst, methane, liquefied petroleum gas, associated petroleum gas, preferential CO methanation DOI: 10.32737/2221-8688-2019-2-193-204

1. Introduction

In the most R&D works on fuel cell (FC) based electrochemical generators in the past 20-25 years, it was assumed that by the beginning of mass introduction of these devices, the fuel (hydrogen) supply infrastructure and logistics would already have been created, including sufficient network of refueling stations and hydrogen storage facilities. The expectations appeared overly optimistic. Even currently existing and actively developing natural gas (raw material for about 60% of global hydrogen production) infrastructure lags behind the mass needs. In this regard, the world research centers involved in the development of FC-based power units focus special attention on conventional widely available high energy density fuels, such as natural gas, liquefied petroleum gas, gasoline, aviation kerosene, diesel fuel, methanol, ethanol, dimethyl ether. Alternative fuels (ammonia,

dimetoxymethane, formic acid, formaldehyde)

are considered as well. It should be noted that among currently available hydrogen storage solutions, hydrocarbon fuel has the highest content of hydrogen per unit volume [1].

Hydrocarbon fuels cannot be oxidized directly on the FC anode at low temperatures, while at higher temperatures they often cause electrode coking and FC damage. Therefore, the fuel is first converted to hydrogen-rich gas (or syngas) in a fuel processor, and the obtained hydrogen is then oxidized in FC [2]. Depending of FC type, hydrogen-rich gas with varying content of CO and CO2 can be used as a fuel. Sulfur-containing compounds are detrimental for all types of fuel cells and must be removed from the fuel during production stage at the refinery or before feeding to fuel processor.

Hydrogen-rich gas can be produced by the processes of steam reforming (SR), partial oxidation (PO), or their combination -autothermal reforming (ATR) of fuel [3-10].

SR of hydrocarbons and oxygenated compounds at temperatures of 300 to 800°C in water excess (molar ratio H20/C = 1.1-3) is the most effective process which provides the highest hydrogen yield. However, it requires a lot of heat for the reaction, water vaporization, steam heating, and needs water conditioning system. Therefore, SR is mainly suggested for use in stationary applications. Integration of endothermic and exothermic processes, which take place in the fuel processor and FC stack at the stages of hydrogen-rich gas production and utilization, facilitate the reaching of high efficiency of FC-based power units. The PO and ATR processes are intended mainly for portable and autonomous applications, for which the priority issues are compactness, water-independence and rare service. Syngas produced by fuel SR, PO and ATR can be used directly for solid-oxide fuel cells (SOFC) feeding.

For using in proton exchange membrane fuel cells (PEMFC), syngas must be deeply purified from CO. For this purpose, CO water gas shift (WGS) reaction is applied. Besides hydrogen, the outlet gas usually contains (vol.%) ~20 CO2, ~10 H2O and ~0.5-2 CO. It can be used for feeding high-temperature PEMFC (operating temperature 160-190 0C), but inappropriate for low-temperature PEMFC (80-120 0C) because CO is the anode poison. CO must be removed from the reformat to the level of <10 ppm. Currently, catalytic preferential (selective) methanation (PreMeth) of CO is considered as one of the most promising processes for deep purification of reformate from CO [11]. The process requires no additional reagents. The presence of CO2, water vapor and methane in the outlet gas has no negative effect on PEMFC operation. However, the best CO PreMeth performance is reached only in the presence of highly active and selective

catalysts. Otherwise, accompanying highly exothermic reaction of CO2 methanation will cause considerable hydrogen losses, catalyst overheating and unstable operation of the fuel processor.

Being highly exothermic, the process of partial oxidation requires no external heating that allows minimization of auxiliary equipment to provide stable operation of the reformer. However, high thermal effect of the reaction implies strong requirements for the catalyst stability and thermal conductivity, and resistance to the active component sintering.

In our earlier works we suggested to use highly heat-resisting (up to 1350°C) FeCrAl alloy materials in a form of meshes and foams as a heat-conducting structural carrier of the catalysts for the reactions of complete and partial oxidation of gaseous hydrocarbons, and steam and autothermal reforming of diesel fuel [12-16]. The applied approach appeared universal and technologically advanced. It allowed manufacturing catalytic modules of various sizes and geometry for all steps of hydrogen-rich gas production in a fuel processor.

The present paper demonstrates the advantages of FeCrAl alloy wire mesh for the production of structured catalytic modules with high heat-conductivity, effective masstransfer and high surface area. The processes of partial oxidation of natural gas (NG), liquefied petroleum gas (LPG), associated petroleum gas (APG) and dimethyl ether (DME) to synthesis gas over the same catalyst were performed to demonstrate the feasibility of multi-fuel concept which suggests the using of various fuels for FC-based power units feeding. The developed procedure was also adapted for the synthesis of structured catalysts for deep CO removal from hydrogen-rich gas mixtures.

2. Experimental part

2.1 Catalyst preparation and characterization

Composite oxide Ce0.75Zr0.25O2-5 (CZ) with fluorite structure was purchased from Ecoalliance Ltd. (Russia). CZ was prepared by

direct precipitation described in detail elsewhere [17]. Solutions of Ce3+ and ZrO2+ nitrates were combined in fixed proportions resulting in concentrations of components totaling 0.1 g cm-3. The composite oxide was

precipitated from the solution by adding 25% aqueous ammonia dropwise under permanent stirring. The mixture was held at 100 °C and atmospheric pressure for 120 h. Thereafter, the precipitate was filtered and repulped in isopropanol. The resulting slurry was dried at 100 °C for 12 h and calcined at 500 °C for 2 h, and then at 800 °C for 1 h. The synthesized CZ material was crashed in an agate mortar, and the 0.2 - 0.5 mm fraction was separated for further catalyst synthesis. Properties of CZ are described in detail elsewhere [17]. CZ specific surface areas (Sbet) was 69 m2 g-1.

Supported Rh/CZ catalyst with metal loading of 1 wt.% was prepared by a sorption-hydrolytic deposition technique according to [18] and described in detail elsewhere [19]. The main idea of this technical approach is to form a solution of "metal complex + alkaline agent" which is metastable at the given conditions (temperature, concentrations) with respect to homogeneous metal hydroxide precipitation. This is due to the kinetic inertness of the metal complexes for ligand exchange [20]. As the support surface accelerates heterogeneous nucleation and growth of metal hydroxide particles [21], addition of support to the reagent mixture initiates hydrolysis which leads to uniform depositing of the metals over the support surface. For this purpose, a solution of RhCl3 was mixed with Na2CO3 in a molar ratio of Na/Cl = 1. The mixture was then brought into contact with CZ support at 75 °C. After the end of the deposition procedure (i.e., when no reaction of the solution with NaBH4 was observed), the sample was thoroughly rinsed with hot water by decanting, dried and reduced in a hydrogen flow at 250 °C for 30 min. SBET of the prepared catalyst was close to

that of the support. According to TEM analysis, the Rh particles in the catalyst existed on the support surface predominantly in a form of 1 - 2 nm clusters in good agreement with the CO chemisorption data. Prepared Rh/CZ catalyst was tested in DME partial oxidation.

FeCrAl alloy wire mesh (grid cell size is 0.5 mm, wire diameter is 0.25 mm) purchased from "NPO Souznichrom" Inc. (Russia) was used as a substrate for catalyst deposition. The main steps for the catalytic module preparation procedure were previously described in detail [16].

Honeycomb - type module consisting of corrugated and flat FeCrAl alloy wire meshes twisted in an Archimedes' spiral was coated by a y-Al2O3 protective layer based on preparation technique described in previous research [12]. Then the y-Al2O3 - coating was impregnated by a saturated solution of Ce(NO3)3'6H2O and ZrO(NO3)2*2H2O (Ce/Zr = 3) during 7 sequential steps with intermediate thermal treatment at 800 °C for 5 min and final calcination at the same temperature for 0.5 h. Thus the Ce0.75Zr0.25O2_5-y-Al2O3/FeCrAl structured support was prepared. Finally, the catalytic module was synthesized by deposition of Rh onto structured support by the sorption-hydrolytic deposition technique described above. For simplicity, Rh/Ce075Zr025O2_5-y-Al2O3/FeCrAl wire mesh catalyst (futher labeled as Rh/CZ-Al/FCA) was tested in partial oxidation of NG, LPG, their mixture modeling APG, and DME. Figure 1a presents the final design of the honeycomb-type composite catalytic module (the size of channel cross -section is ca. 1.7 mm2). Specification of the catalytic module used in the PO experiments is presented in Table 1.

Table 1. Specifications of catalytic modules used in the PO and PreMeth experiments

Module abbreviation Module dimensions, mm Composition, wt.% Total weight,

diameter length Rh Ni Ceo.75Zro.25Ö2-5 CeÜ2 Y- Al2Ü3 g

Rh/CZ-Al/FCA 42 54 0.24 - 6.3 5.5 84.3

Ni/Ce-Al/FCA 8 17 - 1.1 10.0 6.0 0.87

For PreMeth CO catalyst preparation the procedure of depositing the active component on a structured support was adapted. Sandwich-type FeCrAl alloy wire mesh support (Figure 1b) with a fAl2O3 protective coating were impregnated with an aqueous solution of Ce(NO2)3 and calcined at 400 °C to produce the CeO2 film on the support surface. This procedure was repeated several times until the CeO2 amount reached ~10 wt% of the support weight. The CeO2/FCA catalytic module were then

impregnated with an aqueous solution of nickel salts (molar ratio of Ni(NO3)2:NiCl2=4, calculated Ni loading of 10 wt% with regards to CeO2), dried at 110 °C for 2 h and reduced in a flow of 5 vol% H2/He at 400 °C for 2 h. Then the catalytic module was cooled down to room temperature in He flow and taken out to the ambient atmosphere. The active component (Ni, Cl and CeO2) loading in the structured catalyst (further denoted as Ni/Ce-Al/FCA) was ~11 wt% with respect to FCA (See Table 1).

Fig. 1. Images of Rh/CZ-Al/FCA catalytic module twisted in an Archimedes' spiral (a) and Ni/Ce-Al/FCA sandwich - type catalytic module (b).

Previously, the as-prepared and used Rh supported over composite ceria - zirconia -alumina FeCrAl alloy wire mesh catalysts (synthesized by the same preparation protocol as described above for the Rh/CZ-Al/FCA honeycomb catalytic module) [13,15,16] and Ni/Ce-Al/FCA sandwich-type catalytic modules [22,23] were characterized by several techniques. It was shown by transmission electron microscopy (TEM) study that the rhodium in the catalysts uniformly distributed over the ceria - zirconia surface. The sizes of Rh clusters were 1 - 2 nm. Several 3 - 4 nm particles were observed as well. In general, these results agreed well with the CO chemisorption data. According to scanning electron microscopy (SEM) study, a layer of y-Al2O3 consisted of columnar or needle -shaped crystals (5 - 10 p,m thick, up to 50 p,m long) evenly covering the surface of FeCrAl alloy wire mesh. The thickness of the y-Al2O3 coating was 30 - 50 p,m. After supporting Rh/ZrxCe1.xO2-5 over y-Al2O3 layer, the surface microstructure has modified. The surface of the crystals became rougher, while

the thickness of the final layer remained the same (30 - 50 p,m). Concentrations of the Rh, Ce, Zr, Al, and O elements throughout all areas studied by energy-dispersive X-ray analysis were similar that proves uniform distribution of Rh and ZrxCe1-xO2_5 over the surface of y-Al2O3 crystals, in good agreement with TEM data.

The microstructure of the mesh surface also changed substantially after deposition of CeO2 and Ni over the alumina layer. It was shown [22], that the surface was covered by a continuous layer of small (ca. 1-10 p,m) agglomerates of particles. The thickness of this layer was about 90 p,m. EDX analysis showed that the surface layer of Ni/Ce-Al/FCA contains Ce, Ni, Cl, O and a small amount of Al. The concentrations of these elements in all areas were similar, indicating a uniform distribution of Ni, CeO2 and Cl over the alumina layer on FeCrAl alloy wires. The TEM data were in line with SEM data and confirmed that the surface structure of the Ni/Ce/FCA catalyst was similar to that of the powdered chlorinated nickel ceria catalysts

studied in detail in [24].

Thus, the different parts of the Rh/CZ-Al/FCA and Ni/Ce-Al/FCA composite catalysts have their own functionality. Metal wire mesh structure provides high heat conductivity and mechanical strength for the whole module. Alumina intermediate layer in the form of columnar or needle-shaped crystals chemically bonded with the FeCrAl alloy substrate prevents its degradation in the reaction environment. At the same time, such structure provides elasticity of the catalytic coating and preserves secondary Ce1-xZrxO2-s layer with Rh or CeO2 layer with Ni particles in the dispersed state.

The carbon formation in the Rh/CZ pellets and Rh/CZ-Al/FCA catalytic module during catalytic activity experiments was quantified by temperature programmed oxidation (TPO) using a quartz fixed bed flow reactor. The gas mixture of 8 vol.% O2 in He was supplied at a rate of 6 L-h-1. The temperature was increased from 20 to 600 °C continuously at a rate of 10 °C-min-1. A mass-spectrometer QMC - 200 (Stanford Research Systems, USA) was used for online measurement of the outlet CO2 concentration.

2.2. PO activity measurements

DME partial oxidation was carried out in a U-shaped quartz reactor (i.d. 4 mm, wall thickness ~1 mm) at 1 atm in the temperature range of 500-700°C. Rh/CZ catalyst pellets with volume of 0.5 cm3(-0.5 g) was placed in

the reactor. All empty space in the reactor before and after catalyst bed was filled by quartz pellets of the catalyst-size fraction (-0.5 mm) to prevent the homogeneous oxidation of DME, which, as shown by separate experiments, takes place in the gas phase under the studied reaction conditions.

Experiments were performed in the mixture DME:02:N2 = 30:15:55 vol.% at gas hourly space velocity (GHSV) of 10,000 h-1. The temperature was controlled by K - type thermocouple in the middle of the catalyst bed. Prior experiments, the catalyst was reduced at 550 °C in 10 vol.% H2 in the N2 flow during 0.5 hour.

The inlet and outlet concentrations of gas mixtures were measured online (with an accuracy of ± 2 rel.%) using a gas chromatograph Chromos GC - 1000 (Chromos, Russia). It had Porapak T and molecular sieves CaA columns and thermal-conductivity (TCD) detector. The main reaction products were H2, C02, CO, CH4 and H2O. Formaldehyde, methanol and C2 hydrocarbons were observed in trace amounts (less than 200 ppm). The carbon balance was controlled with an accuracy of ± 3 rel.%.

The catalyst performance in the DME partial oxidation was characterized by the outlet concentrations of the main reaction products, DME (XDME) and O2 (XO2) conversion, which were calculated by the following equations:

C0 - C x

DME DME A

XDME (%)-■

C0

N

C

C

x 100 (1)

DME

C0 - C

rt r

Xo2 (%) = ■

C0

N

C

C0

' r.

- x 100

(2)

2

2

2

2

2

0

Rh/CZ-Al/FCA catalytic module was atmospheric pressure, GHSV of 40,000-50,000 tested in a stainless steel fuel reformer under h-1 and the following feed gas compositions:

- NG PO, vol.%: 59,6 N2, 16.0 O2, 21.8 CH4, 1.4 C2H6, 0.5 C3H8, 0.6 Ar.

- LPG PO, vol.%: 70,3 N2, 18.9 O2, 8,1 C3H8, 1,9 C4H10, 0.6 Ar.

- APG PO, vol.%: 63,3 N2, 17.0 O2, 11.5 CH4, 0.6 C2H6, 5.4 C3H8, 1.3 C4H10 0.8 Ar. DME PO, vol%: 57.3 N2, 15.4 O2, 26.5 (CH3)2O, 0.7 Ar

The temperature was controlled by K -type thermocouples at the inlet and outlet of the catalytic module. Prior experiments, the catalyst was reduced at 700 °C in 5 vol.% H2 in the N2 flow.

The inlet and outlet concentrations of gas mixtures were measured online (with an accuracy of ± 2 rel.%) using a gas chromatograph Chromos GC - 1000 (Chromos, Russia). It had Porapak Q and molecular sieves CaA columns and flame ionization (FID) and thermal-conductivity (TCD) detectors. A moisture trap was used for water removal from the gas mixture before analysis. During the experiments, the carbon balance was controlled with an accuracy of ± 3 rel.%. N2, H2, CO, CO2, CH4 concentrations were monitored and compared with the equilibrium values, which were calculated using the HSC Chemistry 7.0 software by minimization of the Gibbs free energy of the system.

2.3 PreMeth CO activity measurements

Preferential methanation of CO in the presence of CO2 was carried out in a U-shaped quartz reactor (i.d. 8 mm, wall thickness ~1 mm) at 1 atm. The sandwich -type Ni/Ce-Al/FCA catalytic module was placed in the reactor. Quartz-wool filters were placed before and after the catalyst. The temperature was measured by a K-type thermocouple (diameter 0.1 mm) placed approximately in the middle of

the catalyst bed. Catalytic experiments were performed using a hydrogen-rich gas mixture of the following composition (vol%): 1.0 CO, 65 H2, 10 H2O, 20 CO2 with He as balance gas, and at GHSV of 29,000 cm3 gcat -1 h-1 with regard to the weight of the deposited Ni(Cl)/CeO2 layer. Initially, the catalyst was heated to 170°C in a He gas flow, and then reduced in the reaction mixture at 400°C for 1 h; after that the temperature was decreased and the experiment was started. The experiment was repeated in several cycles of increased/decreased temperatures. Catalytic characteristics of Ni/Ce/FCA were reproduced in several cycles. The total duration of the tests was ca. 100 h. The inlet and outlet concentrations were determined using a Chromos GC-1000 chromatograph (Chromos, Russia) equipped with a methanation reactor unit and flame-ionization detector. Methane was the main reaction product. The detection limit of CO and CO2 was 1 ppm. The catalyst performance was characterized by the outlet concentrations of CO, CH4 ([CO]out and [CH4]out) and CO selectivity (SCO), which was calculated by the following equation:

fin _ f0ut

Sco(%) = -10°,

F

(3)

ch4

where FCinO - the inlet molar flow rate of CO; F^ andF°u - the outlet molar flow rate of

CH

CO and CH4.

4-50 500 550 600 650 700 750 Temperature, °C

Fig. 2. Effect of temperature on DME conversion and H2, CO2, CO, CH4 concentrations in DME partial oxidation over Rh/CZ catalyst. Points and solid lines - experimental data, dashed lines -equilibrium concentrations.

2. Results and discussion

As Fig. 2 shows, DME conversion on Rh/CZ catalyst reached 100% in the whole studied temperature interval 500-700°C. It is seen that the outlet concentrations of H2, CO, CO2 and CH4 attained respective equilibrium values at temperatures above 630oC. The catalyst demonstrated stable operation and, as proved by TPO experiments, high coking-resistance.

At temperatures below 600°C, the outlet concentrations of DME PO reaction products deviated from the equilibrium values. The gross reaction of partial oxidation of DME, which is described by equation (4), is not elementary:

(CH3)2O + 1/2O2 = 3H2 + 2CO (4) In the frontal part of the catalyst most likely dominate the reactions of DME complete oxidation (5) and decomposition (6): (CH3)2O + 3O2 = 2CO2+3H2O (5) (CH3)2O = H2+ CO+CH4 (6)

The reaction of DME SR (7) can be

realized further along the catalyst layer:

(CH3)2O + 3H2O = 6H2 + 2CO2 (7) Low concentration of CO and high concentration of CH4 in the outlet gas allow suggestion that the reaction of CO methanation (8) accompanies the above reactions (5-7): CO + 3H2 = CH4 +H2O (8)

Combination of all these reactions seems to explain adequately the observed deviation of the outlet product concentrations from respective equilibrium values at temperatures below 600 °C (Fig. 2).

Structured Rh/CZ-Al/FCA catalyst showed excellent catalytic performance in DME PO as well. As Table 2 shows, DME conversion was 100% even at GHSV = 45,000 h-1, which 4.5-times exceeded respective value in experiments with catalyst pellets. The structured catalyst syngas (H2 + CO) productivity was more than 3 m3/h. This value is sufficient for a 3 kWe SOFC feeding.

Table 2. Inlet conditions and experimental results of NG, LPG, APG and DME partial oxidation

with Rh/CZ-Al/FCA catalytic module.

Reaction conditions Type of fuel

NG LPG APG DME

GHSV, h-1 40.000 50.000 40.000 50.000 40.000 50.000 45.000

T inlet C 800 800 950 960 860 870 840

NG flow rate, L/min 11.8 14.7 - - 6.1 7.6 -

LPG flow rate, L/min - - 4.9 6.1 3.3 4.1 -

DME flow rate, L/min - - - - - - 15

Air flow rate, L/min 38.1 47.6 44.9 56.2 40.5 50.6 41.5

Fuel conversion, % 99 98 99 99 99 99 100

Syngas productivity , m3/h 2.5 3.3 2.6 3.1 2.6 3.1 3.2

Outlet gas mixture composition (dry basis), vol%

H2 32.7 32.9 27.2 26.1 31.1 30.6 33.4

O2 0 0 0 0 0 0 0

N2 43.9 43.8 47.6 49.1 45.2 45.6 34.8

CO 16.1 15.9 19.2 19.4 16.9 17.1 22.7

CO2 2.7 2.8 2 1.9 1.9 2.1 3.6

CH4 * . . . . . 2.2 2.4 0.03 0.07 1.2 1.4 1.6

"Calculated on H2 + CO

In our earlier work [16], partial a smaller size (2.5-time smaller than in this oxidation of NG and LPG was performed using work). It was shown that at temperatures 820 structured Rh/CZ-Al/FCA catalytic module of (NG PO) and 950 °C (LPG PO) at the surface

of the catalyst module inlet the measured H2, CO, CO2, CH4, and N2 concentrations were constant and still closed to equilibrium values at GHSV up to 100,000 and 80,000 h-1, respectively. The concentrations of ethane and propane slightly increased from trace amount to 0.02-0.05 vol.% with increasing space velocity. The reformate gas contained high concentrations of H2 and CO and, thus, can be used for SOFC feeding. The maximum observed syngas (CO + H2) productivity per catalyst volume was ~60 m3Lcat-1h-1 (STP). These results showed high performance of composite catalyst, which was achieved, most likely, due to the high heat conductivity of the FeCrAl alloy support. The difference between the hottest and coldest parts of the catalytic module did not exceed 140 °C. Good thermal properties of the catalyst helped to minimize the possibility of local hot spot formation,

catalyst overheating, sintering of the active catalytic phase, and thus prevented catalyst deactivation.

The data presented in Table 2 prove that the experiments with Rh/CZ-Al/FCA catalytic module agree with the previous results. Partial oxidation of NG, LPG, and surrogate APG provided complete conversion of the initial fuel with both GHSV values of 40,000 and 50,000 h-1. The outlet product concentrations were close to the equilibrium ones. According to our calculations, the structured catalyst provided more than 3 m3/h syngas (H2+CO) productivity for all the studied fuels under similar reaction conditions that makes it highly promising for using in a multi-fuel reformer. The TPO experiment with the used Rh/CZ-Al/FCA composite catalyst revealed no carbon formation during NG, LPG and APG PO.

Temperature, °C

Fig. 3. The temperature dependence of the outlet concentration of CO (a), CH4 (b) and CO selectivity (c) for the preferential CO methanation over Ni/Ce-Al/FCA.

Fig. 3 illustrates the performance of Ni/Ce-Al/FCA catalyst in PreMeth CO reaction. It is seen that the catalyst provided the outlet CO concentration below 10 ppm at high selectivity (Fig. 3 a,c). The presence of H -Al2O3 had no negative effect on the Ni/Ce-Al/FCA catalyst performance as we shown earlier [25] by comparison with Ni/CeO2 catalyst pellets. Alumina acted as a structural promoter for the deposited Ni/CeO2 layer. At temperatures below 280 °C, the reaction of CO2 methanation was suppressed (Fig.3b), as

demonstrated by the plateau of the temperature dependencies of the outlet CH4 concentration.

As the temperature exceeded 280 °C, the contribution of CO2 methanation became obvious. As shown in our earlier works [22,26,27], chlorine species CeOCl on the surface of the nickel-ceria catalysts inhibited the CO2 methanation reaction. The structured Ni/Ce-Al/FCA provided CO cleanup to a level of <10 ppm in the reformate gas containing 20 vol.% CO2 in the temperature interval of 230270 °C with a selectivity Sco > 70%.

10-

E œ œ

13

_o

p.

■ 250°C ! 250°C

reformate reformate i i

• • i ■ !• *

Experiments under various reaction conditions i i i ■ ★

■ ★ i , i

★ !★ I I . i I

20 40 60 80

TOS, h

100

-90

-80 _c

' C

-70 </$ -60 50

100

0

0

Fig. 4. CO outlet concentration (★) and selectivity (•) as a function of time on stream in preferential CO methanation over Ni/Ce-Al/FCA at 250°C.

Structured Ni/Ce-Al/FCA catalyst was stable under PreMeth CO conditions. It is seen (Fig. 4) that during the first 13 h Ni/Ce-Al/FCA provided deep CO removal ([CO]out <10 ppm) at Sco ~ 80%. During the next ~60 h the catalyst was tested under varying conditions (CO content, temperature, space velocity). Then the experiment was returned to initial conditions in order to check the catalyst

stability. Clearly, even after 80 h on-stream Ni/Ce-Al/FCA demonstrated stable reproducible characteristics: [CO]out < 10 ppm at SCO ~ 80%. No damage or loses of the catalyst layer were identified after ca. 100 h of catalytic testing, indicating a reasonably good adhesion of the catalytic coating onto the FeCrAl alloy wire mesh.

3. Conclusions

Honeycomb Rh/CZ-Al)/FCA wire mesh composite catalytic module was designed and manufactured. It demonstrated stable operation and excellent catalytic performance in multifuel partial oxidation

reformer with natural gas, liquefied petroleum gas, associated petroleum gas and dimethyl ether. It provided equilibrium product distribution at GHSV 40,000-50,000 h-1. The

syngas (CO + H2) productivity was sufficient to feed a 3 kWe solid oxide fuel cell.

A structured Ni/Ce-Al/FCA wire mesh catalyst was developed for the reaction of preferential CO methanation in the presence of CO2. In a hydrogen-rich gas mixture of the

composition (vol.%): 1.0 CO, 65 H2, 10 H2O, 20 CO2, He - balance, GHSV = 29,000 cm3-gcat"1h"1, the catalyst reduced the CO concentration to a level of <10 ppm in the temperature interval of 230-270°C with a selectivity >70%.

Acknowledgements

In the part of hydrogen-rich gas mixtures purification from CO, this work was supported by Ministry of Science and Higher Education of the Russian Federation (Project AAAA-A17-117041710088-0).

In the part of partial oxidation of hydrocarbons and dimethyl ether, V.A.Sobyanin, S.D. Badmaev, A.A. Pechenkin, P.V. Snytnikov would like to acknowledge the financial support from the Russian Science Foundation, Project № 17-79-3007.

References

1. Shekhawat D., Berry D.A., Spivey J.J. Introduction to fuel processing, in: Fuel Cells Technol. Fuel Process., Elsevier. 2011. pp. 1-9.

2. Song C. Fuel processing for low-temperature and high-temperature fuel cells. Challenges, and opportunities for sustainable development in the 21st century. Catal. Today. 2002, vol. 77, no. 12, pp. 17-49.

3. Navarro R.M., Peña M.A., Fierro J.L.G. Hydrogen production reactions from carbon feedstocks: fossil fuels and biomass. Chem. Rev. 2007, vol. 107, no. 10, pp. 3952-3991.

4. Li D., Li X., Gong J. Catalytic reforming of oxygenates: state of the art and future prospects. Chem. Rev. 2016, vol. 116, no. 19, pp.11529-11653.

5. Xu X., Li P., Shen Y. Small-scale reforming of diesel and jet fuels to make hydrogen and syngas for fuel cells: a review. Appl. Energy. 2013, vol. 108, pp. 202-217.

6. Lindermeir A., Kah S., Kavurucu S., Mühlner M. On-board diesel fuel processing for an SOFC-APU—Technical challenges for catalysis and reactor design. Appl. Catal. B Environ. 2007, vol. 70, no. 1-4, pp. 488-497.

7. Sengodan S., Lan R., Humphrey J.s, Du D., Xu W., Wang H., Tao S. Advances in reforming and partial oxidation of hydrocarbons for hydrogen production and fuel cell applications. Renew. Sustain.

Energy Rev. 2018, vol. 82, pp. 761-780.

8. Abatzoglou N., Fauteux-Lefebvre C., Review of catalytic syngas production through steam or dry reforming and partial oxidation of studied liquid compounds. Wiley Interdiscip. Rev. Energy Environ. 2016, vol. 5, no. 2, pp. 169-187.

9. Bae J., Lee S., Kim S., Oh J., Choi S., Bae M., Kang I., Katikaneni S.P. Liquid fuel processing for hydrogen production: a review, Int. J. Hydrogen Energy. 2016, vol. 41, no.44, pp. 19990-20022. ^

10. Navarro Yerga R.M., Alvarez-Galvan M.C., Mota N., Villoria de la Mano J.A., Al-Zahrani S.M., Fierro J.L.G., Catalysts for hydrogen production from heavy hydrocarbons. ChemCatChem. 2011, vol. 3, no. 3, pp. 440-457.

11. Snytnikov P.V., Zyryanova M.M., Sobyanin V.A. CO-cleanup of hydrogen-rich stream for LT PEM FC feeding: catalysts and their performance in selective CO methanation. Top. Catal. 2016, vol. 59, no.15-16, pp. 1394-1412.

12. Porsin A.V., Kulikov A.V., Rogozhnikov V.N., Serkova A.N., Salanov A.N., Shefer K.I. Structured reactors on a metal mesh catalyst for various applications, Catal. Today. 2016, vol. 273, pp. 213-220.

13. Shoynkhorova T.B., Rogozhnikov V.N., Simonov P.A., Snytnikov P.V., Salanov A.N., Kulikov A.V., Gerasimov E.Y., Belyaev V.D., Potemkin D.I., Sobyanin V.A., Highly dispersed Rh/Ce0.75Zr0.25O2-

s-y-Al2O3/FeCrAl wire mesh catalyst for autothermal n-hexadecane reforming. Mater. Lett. 2018, vol. 214, pp. 290-292.

14. Rogozhnikov V.N., Snytnikov P.V., Salanov A.N., Kulikov A.V., Ruban N.V., Potemkin D.I., Sobyanin V.A., Kharton

Rogozhnikov V.N., Gerasimov E.Y., Salanov A.N., Belyaev V.D., Sobyanin V.A. From alumina modified Rh/Ce0.75Zr025O2_5 catalyst towards composite Rh/Ce0.75Zr0.25O2_s-y-

Al2O3/FeCrAl catalytic system for diesel conversion to syngas. Appl. Catal. B Environ. 2019, vol. 245 pp. 40-48.

16. Shoynkhorova T.B., Rogozhnikov V.N., Ruban N.V., Shilov V.A., Potemkin D.I., Simonov P.A., Belyaev V.D., Snytnikov P.V., Sobyanin V.A. Composite Rh/Zr0.25Ce0.75O2_s-y-Al2O3/Fecralloy wire mesh honeycomb module for natural gas, LPG and diesel catalytic conversion to syngas. Int. J. Hydrogen Energy. 2019, vol. 44, no. 20, pp. 9941-9948.

17. Alikin E.A., Denisov S.P., Danchenko N.M., Rychkov V.N., Volkov A.S., Karpov A.S. Thermally stable composite system Al2O3 - Ce0.75Zr0.25O2 for automotive three way catalysts. Catal. Ind 2013, vol. 5, no. 2, pp. 133-142.

18. Patent №2653360 Russia. Method of catalyst preparation / Simonov P.A., Shoynkhorova T.B., Snytnikov P.V., Potemkin D.I., Belyaev V.D., Sobyanin V.A. 2018.

19. Shoynkhorova T.B., Simonov P.A., Potemkin D.I., Snytnikov P. V., Belyaev V.D. Sorption-hydrolytic deposition for diesel fuel reforming to syngas. Appl. Catal. B Environ. 2018, vol. 237, pp. 237244.

20. Wu L., Schwederski B.E., Margerum D.W. Stepwise hydrolysis kinetics of tetrachloroplatinate(II) in base. Inorg. Chem. 1990, vol. 29, no. 18, pp. 35783584.

21. Kashchiev D., van Rosmalen G.M. Review: nucleation in solutions revisited. Cryst. Res. Technol. 2003, vol. 38, no. 78,

V.V., Rh/9-Al2O3/FeCrAlloy wire mesh composite catalyst for partial oxidation of natural gas. Mater. Lett. 2019, vol. 236, pp. 316-319.

15. Shoynkhorova T.B., Snytnikov P.V., Simonov P.A., Potemkin D.I.,

pp.555-574.

22. Konishcheva M.V., Snytnikov P.V., Rogozhnikov V.N., Salanov A.N., Potemkin D.I., Sobyanin V.A. Structured Ni(Cl)/CeO2/n-Al2O3/FeCrAl wire mesh catalyst for selective CO methanation. Catal. Commun. 2019, vol. 118, pp. 2529.

23. Konishcheva M.V., Potemkin D.I., Snytnikov P.V., Sobyanin V.A. The influence of CO, CO2 and H2O on selective CO methanation over Ni(Cl)/CeO2 catalyst: On the way to formic acid derived CO-free hydrogen. Int. J. Hydrogen Energy. 2019, vol. 44, no.20, pp. 9978-9986.

24. Konishcheva M.V., Potemkin D.I., Snytnikov P.V., Stonkus O.A., Belyaev V.D., Sobyanin V.A. The insights into chlorine doping effect on performance of ceria supported nickel catalysts for selective CO methanation. Appl. Catal. B Environ. 2018, vol. 221, pp. 413-421.

25. Zyryanova M.M., Snytnikov P.V., Amosov Y.I., Venyaminov S.A., Golosman E.Z., Sobyanin V.A., Selective methanation of CO in the presence of CO2 in hydrogen-containing mixtures on nickel catalysts. Kinet. Catal. 2010, vol. 51, no. 6, pp. 907-913.

26. Konishcheva M.V., Potemkin D.I., Badmaev S.D., Snytnikov P.V., Paukshtis E.A., Sobyanin V.A., Parmon V.N. On the mechanism of CO and CO2 methanation over Ni/CeO2 Catalysts. Top. Catal. 2016, vol. 59, no. 15-16, pp. 1424-1430.

27. Zyryanova M.M., Snytnikov P.V., Gulyaev R.V., Amosov Y.I., Boronin A.I., Sobyanin V.A., Performance of Ni/CeO2 catalysts for selective CO methanation in hydrogen-rich gas. Chem. Eng. J. 2014, vol. 238, pp. 189-197.

КАТАЛИЗАТОРЫ И КАТАЛИТИЧЕСКИЕ ПРОЦЕССЫ ДЛЯ ПОЛУЧЕНИЯ ВОДОРОДСОДЕРЖАЩЕГО ГАЗА ДЛЯ ПИТАНИЯ ТОПЛИВНЫХ ЭЛЕМЕНТОВ

12 12 3 12 13 12

С.Д. Бадмаев ' , В.Д. Беляев ' ' , М.В. Конищева ' , А.В. Куликов ' , А.А. Печенкин ' , Д.И. Потемкин1'2, В.Н. Рогожников1, П.В.Снытников12'3, В.А.Собянин1

1 Институт катализа СО РАН 630090, Новосибирск, пр. Академика Лаврентьева, 5

2 Новосибирский государственный университет 630090, Новосибирск, ул. Пирогова, 2 3ООО «УНИКАТ», 630090, Новосибирск, пр. Академика Лаврентьева, 5

В работе обсуждаются новые подходы по приготовлению структурированных катализаторов и особенности проведения процессов получения водородсодержащего газа из углеводородов и диметилового эфира для питания топливных элементов различных типов, Также представлены результаты исследования структурированного катализатора очистки реформата от СО на примере избирательного метанирования моноксида углерода.

Ключевые слова: парциальное окисление, диметиловый эфир, структурированный катализатор, метан, сжиженный нефтяной газ, попутный нефтяной газ, избирательное метанирование СО

YANACAQ ELEMENTLdRiNDd iSTiFADd OLUNAN HÍDROGEN TdRKÍBLÍ QAZIN ALINMASI ÜQÜNKATALiZATORLAR Vd KATALiTiKPROSESLdR

S.D. Badmayev1'2, V.D. Belyaev1'2, M.V. Koni^geva1'2, A.V. Kulikov1'3 , A.A. Pegenkin1'2, D.I. Potemkin1'2t V.N. Rogojnikov1, P.V. Snitnikov1'2'3t V.A. Sobyanin1

1Rusiya EA-nin Boreskov ad. Kataliz institutu 630090, Novosibirsk, Akademik Lavrentievpr., 5

2 Novosibirsk Dövldt Universiteti 630090,Rusiya, Novosibirsk §dh., Pirogov küg., 2

3 «UNICAT» LLC, 630090, Novosibirsk, Akademik Lavrentiev pr., 5 e-mail: sobyanin@catalysis.ru

Müxtdlif növ yanacaq elementldri ügün karbohidrogenldrddn vd dimetil efirinddn hidrogen tdrkibli qazin alinmasi ügün strukturla§dirilmi§ katalizatorlarin hazirlanmasinin yeni üsullari müzakird olunur. Eldcd dd riformatin karbon monooksidddn tdmizldnmdsi ügün strukturla§dirilmi§ katalizatorlar carbon monooksidin metanla§ma reaksiyasinda öyrdnilib. Agar sözlw. yanacaq elementldri, dimetil efiri, strukturla§dirilmi§ katalizatorlar, metan, karbon monooksidi