Al2O3 Catalyst

0 downloads 0 Views 219KB Size Report
Abstract—The specifics of CO hydrogenation over a 5%Ru/Al2O3 catalyst in a flow ... surface ignition (CSI) mode—either by heating the catalyst in the reaction ...

ISSN 00231584, Kinetics and Catalysis, 2013, Vol. 54, No. 2, pp. 225–232. © Pleiades Publishing, Ltd., 2013. Original Russian Text © M.A. Kipnis, P.V. Samokhin, E.A. Volnina, 2013, published in Kinetika i Kataliz, 2013, Vol. 54, No. 2, pp. 235–242.

Carbon Monoxide Hydrogenation in the Mode of Ru/Al2O3 Catalyst Surface Ignition M. A. Kipnis*, P. V. Samokhin, and E. A. Volnina Topchiev Institute of Petrochemical Synthesis, Russian Academy of Sciences, Leninskii pr. 29, Moscow, 119991 Russia *email: [email protected] Received July 18, 2012

Abstract—The specifics of CO hydrogenation over a 5%Ru/Al2O3 catalyst in a flow reactor at a pressure of 1.5 MPa has been considered. The feed gas mixture has been composed of (vol %) 30.5 CO, 2.3 CO2, 65 H2, and N2 as the rest. The CO methanation reaction readily passes to the externaldiffusion regime—catalyst surface ignition (CSI) mode—either by heating the catalyst in the reaction medium or by replacing H2 with the reactant gas having a temperature above the critical ignition temperature. On passing to the CSI mode, the temperature at the entrance to the catalyst bed and the methane content at the reactor outlet abruptly increase, the yield of CO2 produced via the watergas shift reaction increases, and the CO content drops to zero. Under the CSI regime, temperature oscillations with a period of 3–5 min and an amplitude of ~3°C are observed, which are sustained during catalyst cooling until the extinction of the reaction. A comparison of the product compositions at the reactor outlet in the cases of the “thick” (20 mm) and “thin”(3 mm) catalyst bed has shown that the reverse water–gas shift, an endothermic reaction, occurs in lower, colder layers of the thick bed. As a result, the extinction of the reaction is faster in the thick than in the thin bed. Methanation of CO is accompanied by the Fischer–Tropsch reaction: a variety of carbon compounds are formed with their yield being decreased on passing to the CSI mode. DOI: 10.1134/S0023158413020080

Reactions involving CO hydrogenation play an important part in the modern chemical industry. For example, a mixture of CO and Н2 (synthesis gas) is used for methanol production, hydroformylation (oxo process), and Fischer–Tropsch (FT) synthesis [1]. The methanation reaction (I) is of great practical importance as it is used for removal of CO from hydro genrich gases designed for fuel cells [2–5] or CO con version to methane [6]: CO + 3H2 = CH4 + H2O. (I) One of the main features of hydrogenation reac tions is their high exothermicity. These reactions are characterized by occurrence in the externaldiffusion regime, in which the catalyst activity slightly depends on temperature. The transition of a reaction to this regime was called catalyst surface ignition (CSI) by FrankKamenetskii [7]. The CSI phenomenon was observed in the reac tions of ethylene hydrogenation on a Pd/boehmite catalyst [8]; selective hydrogenation of ethylene and acetylene on Pd/Al2O3 [9]; hydrogenation of para methylstyrene, octene and heptene on Pd or Pt cata lysts [10]; and CO methanation on fused iron catalysts [11, 12]. The possibility of running the FT synthesis in the CSI mode remains an open question. Earlier [13–17], we showed that highly exothermic reactions, such as oxidation of Н2 and selective oxida tion of CO, could be accomplished in the CSI mode.

The specifics of transition to and exit from this regime were studied, as well as the characteristics of the steady state. In addition, we studied the influence of strong adsorption of CO and О2 on the catalyst bed tempera ture in the CSI regime. The approach developed in the cited studies was applied by us to the methanation reaction on a ruthe nium catalyst. The choice of ruthenium was due to its high activity in the CO methanation and hydrocar bon hydrogenation reactions [18, 19] and the recently increased interest in its use for these reactions [3, 20–30]. In this paper, we consider the specific features of CO hydrogenation on a Ru/Al2O3 catalyst in a flow reactor under conditions that favor the establishment of the CSI regime. EXPERIMENTAL As a support, we used a 0.2–0.315 mm γAl2O3 fraction obtained by crushing commercial extrudates manufactured by the Ryazan refinery (A64k brand; specific surface area, 200 m2/g), which was calcined in air at 500°С for 2 h prior to deposition of the active component. The ruthenium source was the salt Ru(OH)Cl3 (Aurat, Russia). A portion of the support was placed in an aqueous solution of the salt at room temperature in an amount required for obtaining the catalyst with a Ru content of 5 wt %; held over 20 h;

225

226

KIPNIS et al.

and, after decantation, dried for 6 h at 120°С. The fin ished catalyst was reduced in a hydrogen stream at 350°С for 2 h in the reactor. Catalytic experiments were performed in the con tinuousflow setup having a stainless steel reactor according to the scheme described in [31] with the dif ference that the effluent gas stream was withdrawn from the reactor through special heated connections. A catalyst sample was placed on a porous quartz frit (H. Baumbach & Co., United Kingdom). The experi ments were made with two samples of 0.15 and 1 g to have a bed height of 3 mm (“thin” bed) or 20 mm (“thick” bed), respectively. The reactor temperature was monitored with two thermocouples of 1 mm diameter, which were enclosed in a metal jacket with a wall thickness of 0.25 mm mounted along the reactor axis and were fixed at heights differing by ~20 mm. This arrangement made it possible to monitor heating at the entrance to and the exit from the catalyst bed at a bed thickness of 20 mm or the gas temperature over the bed and the bed entrance temperature in the case of the thin bed (height 3 mm). Under steadystate conditions in a nitrogen flow, the difference in readings of the thermocouples did not exceed 0.4°С. Hereinafter, the readings of the thermocouples in the oven, at the entrance to the catalyst bed, at the exit from the bed, and in the bed per se (at bed height of 3 mm) are denoted as To, Tent, Tex, and Tbed, respectively. The feed reaction mixture has the following com position, vol %: 30.5 CO, 2.3 CO2, and 65 H2; the rest was N2. The mixture was prepared by mixing gases according to the pressure (error at most 5 rel. %). Since elevated pressure favors reaction (I), experi ments were run at a pressure of 1.5 MPa. The feed mixture was introduced into the reactor through a gas flow controller (Bronkhorst HighTech B.V., the Netherlands; an F121M mass flow meter equipped with an F033C control valve). All the com ponents of the device downstream of the gas flow con troller to the sampling port for online chromato graphic analysis were maintained at a temperature of 135°С. Liquid products at the exit from the hot zone were condensed in traps at 0°С. The dried gas was ana lyzed on a column packed with molecular sieves 13 X and CO2 was determined on a Polisorb a chromato graphic column using a thermal conductivity detector (helium as the carrier gas). To determine organic car bon products, a column with the SP1700 stationary phase and a flameionization detector were also used. The CO conversion (хСО) was calculated as follows: x СО =

0 C СО − C СО K × 100, 0 C СО

(1)

0 where CСО and ССО are the CO concentrations at the inlet and outlet, respectively; K = C N0 2 C N 2 , C N0 2 , and

C N 2 are the inlet and outlet concentrations of nitro gen, respectively. Since the volume of the gas mixture changes during the reaction, it is convenient for discussion of the experimental data to express product yields and amounts of the unconverted reactants in L/h: Vi = СiV, (2) where Vi and Сi are the volume and the concentration of the ith component, respectively, and V is the volume of the dry gas at the outlet. RESULTS AND DISCUSSION Temperature Change in the Thick Catalyst Bed and C1 Product Composition in the CSI Mode Changes in the catalyst bed temperature, CO con version, and product yields in the case of linear heating of the reaction mixture flowing through the 20mm bed are shown in Fig. 1. Elevation of the oven temper ature leads to gradual heating of the catalyst bed, pri marily, at the exit. Beginning from a certain time (~397 min), the amount of heat released at the exit from the bed in the exothermic CO hydrogenation reaction begins to exceed the amount of heat carried away by the gas stream, thereby causing an upward shift in the reaction front in the bed (Fig. 1a). This is accompanied by reaching an almost 100% conversion of CO (Fig. 1b), i.e., the reaction passes to the CSI regime [13]. The methane content at the reactor outlet dramatically increases, and the yield of СО2 by the 409th min increases from 0.12 (initial quantity) to 0.27 L/h (Fig. 1b). An increase in the yield of СО2 is an indication that along with reaction (I), the water gas shift takes place [32] (CO reacts with the water produced by methanation): СО + Н2О = СО2 + Н2.

(II)

Hydrogen released in this reaction, in turn, can also react with CO yielding an additional amount of methane. Therefore, the yield of СН4 after transition of the reaction to the CSI mode reaches ~1.3 L/h (409 min, Fig. 1b), which is greater than ~1.1 L/h, the value corresponding to the consumption of initial H2 via reaction (I) only. The experimentally measured total yield of СН4 and СО2 is 1.57 L/h, which agrees well within the experimental error with the calculated value of 1.61 L/h attainable in the case of complete conversion of CO. Methane is formed only via hydro genation of CO, since СО2 hydrogenation on ruthe nium does not take place in the presence of CO [21, 24, 33]. Note that the yield of methane is insignificant before passing to the CSI regime, although the CO conversion reaches 26% (Fig. 1b, 385 min). As will be shown below, the increase in CO conversion at a low yield of methane is due to the fact that the main CO KINETICS AND CATALYSIS

Vol. 54

No. 2

2013

CARBON MONOXIDE HYDROGENATION T, °C 360 340 320 300 280 260 240 220 200 180 300

T, °C 390

(a) Тent Тex Тo

340

(a)

Тent Тex Тo

290 240 190 340

380

420

t, min xCO, % 100 90 80 70 60 50 40 30 20 10 0 300

227

CH4, CO2, L/h 1.4

(b) xCO CH4 CO2

1.2 1.0 0.8

340

380

140 260 300 340 380 420 460 500 540 580 620 t, min Тent, °C CO, CH4, CO2, L/h (b) 3.0 380 2.0

300

0.6

260

0.4

220

0.2 0 420

180 140 380

t, min Fig. 1. Transition of the reaction to the CSI mode in the case of linear heating of the reactor at a rate of 1°C/min: (a) temperature changes at the entrance to the catalyst bed, at the exit from the bed, and in the oven and (b) changes in the CO conversion and the CH4 and CO2 yields. The bed height is 20 mm and the flow rate is 5 L/h.

2.5

340

Тent, °С 380

Тent CH4 CO2 CO

1.5 1.0 0.5 0

420

460

500 540 t, min

580

620

(c)

379 378 377

hydrogenation products before this moment are car bon compounds other than methane. The data in Fig. 2 show that the reaction can also be made to proceed in the CSI mode by replacing Н2 with the reaction mixture providing that the catalyst temperature does exceed the critical ignition temper ature [13]. Thus, at a temperature below 259°С (after stepwise successive heating to 175 and 219°С), the replacement of Н2 by the reactant gas results in only slight heating of the bed and it is not until 259°С that the bed temperature abruptly rises. The rise is charac terized by the following difference between the tem peratures: Tent = 378°C, Tex =309°C, and To = 259°C (Fig. 2a). Before the ignition, these temperatures coincide in the Н2 medium. The transition of the reac tion to the ignition regime is accompanied by the com plete consumption of CO and the appearance of a large amount of methane (Fig. 2b), as noted above. In the steadystate region of the CSI regime (Tent ≈ 378°C), the yield of СО2 is 0.52 L/h (its content in the feed gas corresponds to 0.18 L/h), which indicates the simultaneous occurrence of the CO methanation and watergas shift reactions. KINETICS AND CATALYSIS

Vol. 54

No. 2

2013

376 375 420 425 430 435 440 445 450 455 460 t, min Fig. 2. Transition of the reaction to the CSI mode and the back transition with the change of the oven temperature: (a) changes in Tent, Tex, and To (replacement of H2 by the reactant gas is marked by solid arrows and the reverse change is marked by dotanddash arrows); (b) changes in Tent and the CO, CO2, and CH4 yields; and (c) Tent oscil lations under the CSI regime. The bed height is 20 mm and the flow rate is 8 L/h.

As the oven temperature is subsequently lowered and the bed is cooled, the yield of CO increases and the methane and СО2 yields decrease, beginning from the 470th min (Fig. 2b), until a temperature of ~250°C is reached in the hotspot at the bed entrance(~606th min, Fig. 2a). Then, the entrance and exit temperatures get close one another and the reaction is quenched. During quenching, the bed tem peratures become close to To (Fig. 2a), the CO and

228

KIPNIS et al.

T, °C 420

(a) Тbed Тg Тo

370 320 270 220 400

440

420

460 t, min

480

CO, CH4, CO2, L/h (b) 3.0

500

520

H2O, vol % 30

2.5

25

2.0

20

CO CO2 CH4 H2O

1.5 1.0

15 10

0.5 0 400

5 420

T, °C 406

440

460 480 t, min (c)

500

0 520 T, °C 350

348 404 346

Тbed Тg 402 480

490

500 t, min

510

344 520

Fig. 3. Transition of the reaction to the CSI mode during linear heating of the reactor at a rate of 1°C/min: (a) changes in Tbed, Tg, and To; (b) changes in the CO, CO2, and CH4 yields and in the water content of the wet gas; and (c) changes in Tbed and Tg after transition of the reaction to the CSI regime. The bed height is 3 mm and the flow rate is 8 L/h.

СО2 yields return to their initial values, and the yield of СН4 drops to zero (Fig. 2b). After transition of the reaction to the CSI mode, temperature oscillations at the bed entrance (Fig. 2c) are observed to have a period of 3–5 min and an amplitude of ~3°С, being retained until the extinction of the reaction with a decrease in To. No Tex oscillation is observed in this case. The nature of the oscillations is not quite clear and calls for further investigation. It may be assumed that their emergence is due to the dis proportionation reaction

T, °C 400 360 320 280 240 200 1 160 120 80 0 40

(a)

3 2

Тbed Тg Тo

80 120 160 200 240 280 320 t, min CO, CH4, CO2, L/h Тbed, °C 400 2.5 (b) 350

2.0 1.5 1.0

CH4 CO2 CO Тbed

300 250

0.5 0 40

200 80

150 120 160 200 240 280 320 t, min

Fig. 4. Changes in the temperatures and the yield of C1 products with a decrease in the oven temperature after transition of the reaction to the CSI regime: (a) Tbed, Tg, and To (arrows mark the times of (1) replacement of H2 by the reactant gas at a flow rate of 8 L/h, (2) the increase in the feed flow rate to 16L/h, and (3) the decrease in the flow rate to 4 L/h) and (b) the yields of CO, CO2, and CH4. The bed height is 3 mm.

2СО = С + CО2, (III) which has a high value of the equilibrium constant at these temperatures [34]. The carbon produced in reaction (III) is seemingly removed by hydrogen or water. Note that oscillations associated with a change in the state of Ru surface (transition from the metal state to the oxidized state and vice versa) have been detected in the oxidation reactions of CO [14, 35, 36] and CH4 [37]. It is also noteworthy that oscillations with a period of about 1 h were observed in the FT reaction over ironcontaining zeolite catalysts in the case of a sufficiently large sam ple (58 mL) [38]. Tsotsis et al. [38] considered the car bon formation and removal reactions in combination with heat and mass transfer processes to be a possible cause of this phenomenon. The obtained data allow for the conclusion that reactions in the CSI mode occur primarily in a small hot zone at the bed entrance. Consequently, substan tially decreasing the size of the catalyst sample; i.e., using the socalled thin bed, we should observe CSI anyway. Indeed, the transition to the CSI mode was observed by the 471st min (Fig. 3a) at the decreased KINETICS AND CATALYSIS

Vol. 54

No. 2

2013

CARBON MONOXIDE HYDROGENATION T, °C 380 360 340 320 300 280 260 240 220 200 180 160 360

(a)

410

460

Tвх, °C 380

Fig. 5. Changes in (a, b) Tent, Tex, and To; (c) the product composition; and (d) the conversion and CO, CO2, CH4 total yield during heating, transition of the reaction to the CSI mode and subsequent cooling. The arrows in Figs. 5c and 5d mark the direction of variation in oven tempera ture. The bed height is 20 mm and the flow rate is 8 L/h.

Tent Тex Тo

510 560 t, min

610

660

710

(b)

370 365 360 355 350 470

CO, L/h 2.8 2.4 2.0 1.6 1.2 0.8 0.4 0 140

510

490 t, min

(c)

CH4, CO2, L/h 2.1

1.4 CO CH4 CO2

0.7

240 290 190 340 Hotspot temperature, °C

loading of catalyst from 1 to 0.15 g (which corresponds to a decrease in the bed height from 20 to 3 mm) under the linear heating conditions. It should be noted that the gas temperature Tg over the catalyst bed as mea sured by the thermocouple mounted 16 mm above the bed level substantially exceeded To during the transi tion to the CSI regime and further in the steady state (Fig. 3a). This difference is due to propagation of heat released in the reaction medium upward the reactor wall and in the thermocouple jacket. In the steadystate CSI regime, the complete con version of CO yielding СН4 and СО2 was observed as the outcome (Fig. 3b), and the appearance of water immediately at the reactor outlet was detected. Thus, as in the case of the thick (20 mm) bed, the watergas shift reaction proceeds along with the methanation of CO. As in the experiments with the 20mm bed (Fig. 2c), temperature oscillations were observed both upon passing to the CSI regime and in the steady state (Fig. 3c).

375

345 450

229

Figure 4 shows the effect of decrease in the oven temperature on the behavior of the thin catalyst bed. After replacement of Н2 by the reaction mixture, the reaction passes to the CSI mode at To = 300°C (Fig. 4a).Lowering To to ~90°C is not accompanied by extinction of the reaction. Moreover, an increase in the feed flow rate from 8 to 16 L/h at this temperature caused a dramatic rise in Tbed, thereby indicating that the CO conversion increases with the increasing load. It was not until the load decreased to 4 L/h that the Peak height, mV 400

0 390

2 300 ΣC1, L/h 2.8

xCO, % 100

(d)

2.6

80

2.4

60

1 200

100 2.2 2.0 1.8 140

40

ΣC1 xCO

20 0

240 290 190 340 Hotspot temperature, °C

KINETICS AND CATALYSIS

Vol. 54

390

No. 2

2013

2

8 14 Elution time, min

20

Fig. 6. Chromatograms of the CO hydrogenation products as recorded (1) before and (2) after the transition of the reaction to the CSI mode. The hotspot temperatures are (1) 282 and (2) 268°C. The bed height is 20 mm and the flow rate is 8 L/h.

230

KIPNIS et al.

reaction was quenched and Tbed became almost equal to To. Whereas CO is completely converted into CH4 and CO2 at Tbed ≈ 396°C, both the CO conversion and the amount of these products begin to decrease after the drop in the bed temperature below ~290°С (Fig. 4b). At this, the total yield of the С1 components remains at a constant level of about 2.5 L/h, which approxi mately corresponds to their initial content. If the reac tant hydrogen is completely consumed for the forma tion of СН4, the yield of the latter should be 1.7 L/h. However, the actual yield of methane is greater than this value at Tbed above ~250°С (Fig. 4b, until the ~250th min), although it decreases to 1.27 L/h when the temperature is further reduced (317 min, Tbed = 178°C). This means that the reactant hydrogen is incompletely consumed at temperatures below 250°C. This conclusion is supported by direct data on the gas composition. At the same time, the reaction is not quenched; that is, a significant drop in catalyst activity cannot be considered the cause of incomplete hydro gen consumption. We believe that a certain role is played by the bed thickness, which is insufficient for the complete conversion of H2 at these temperatures. The decrease in the yield of CO2 produced via reac tion (II) observed after 250 min at temperatures below ~250°С is due to both the temperature drop and the rise in the concentration of hydrogen that is not con sumed in the methanation reaction. A comparison of the Tent values measured during reduction in the oven temperature shows that the reac tion extinction is faster in the thick, rather than the thin bed (cf. Figs. 2a and 4a). Indeed, the extinction in the thick bed starts when Tent reaches 250°С, as noted above, whereas it does not occur in the thin bed even when the bed entrance temperature decreases to 180°С. It may be assumed that the gas composition in the thickbed cross section below 3 mm is the same as that at the exit from the thin bed. In the case of the thick bed, this gas further gets into the underlying layer in which additional reactions between the components can take place. At hotspot temperatures above ~350°С (this spot occurs at the bed entrance in the thick bed or entirely occupies the thin bed), the total yield of the С1 com ponents does not depend on the bed thickness. How ever, as the thick bed temperature decreases, the yield of СО2 becomes lower and the yield of CO becomes higher than in the thin bed at the same hotspot tem perature (cf. Figs. 2b and 4b). The decrease in the СО2 yield and the increase in the CO yield are presumably due to the occurrence of reverse water–gas shift reac tion (II), which leads to a reduction in Н2 concentra tion and a growth in CO and Н2О concentrations. The cause of the reverse reaction (II) in the lower layers of the thick bed is easy to comprehend by considering the

temperature distribution in the bed (Fig. 2a). As has been noted above, the temperature at the exit from the bed is as low as ~309°С even after the ignition when the bed entrance temperature is ~378°С. Such a drop in temperature is due to heat withdrawal across the reactor walls. Therefore, a reduction in Tent leads to a further decrease in temperature downstream in the bed and, consequently, displaces the equilibrium toward the reverse reaction. While the direct reactions (I) and (II) are highly exothermic, the reverse reaction (II) is strongly endot hermic and is facilitated by cooling the catalyst. We believe that it is for this reason that the extinction of the reaction in the thick bed occurs at a higher oven temperature than in the thin bed. Formation of Hydrocarbons before and after Reaction Transition to the CSI Mode As has been shown above, the main CO conversion products of the reaction in the CSI mode are methane and СО2. At the same time, other products are formed as well before the transition of the reaction to the CSI regime. Figure 5 shows changes in (a, b) temperature, (c) product composition, and (d) total yield of C1 compo nents (CO, CO2, CH4) during heating, transition to the CSI mode, and subsequent catalyst (20mm bed) cooling in the reaction mixture. The reaction in the CSI regime proceeds in the steadystate mode even after reduction in temperature of the external heater (oven). The data in Fig. 5a show that To is maintained at the level of 200°С over the time range of 617–665 min during cooling. After tran sition to the CSI regime (5a), temperature oscillations at the bed entrance are observed, as has been already noted above (see Figs. 2c, 3c), and they are retained when To is decreased (Fig. 5b). The transition of the reaction to the CSI regime and subsequent cooling are accompanied by a hyster esis loop in the temperature dependence curves for the concentration of products, primarily, CO and СН4 (Fig. 5c) in the hotspot (the hotspot occurs at the exit from the bed prior to ignition and at the entrance after the ignition). When the steady state is reached, a small break in the curves is observed at a temperature of ~280°С (Figs. 5c, 5d). The CO conversion at the time of ignition abruptly increases from 29 (~270°С) to 99.5% (~375°С, Fig. 5d), but it monotonically decreases with decreasing temperature owing to the reverse water–gas shift reaction. However, as the hotspot temperature decreases to ~270°С, the CO conversion does not fall to the level corresponding to this value of temperature if attained by heating, but amounts to ~52%. A hysteresis loop is also observed in the temperature dependence curve for the total yield of the С1 components (ΣС1): ΣС1 decreases from 2.64 to 2.06 L/h as the hotspot temperature increases to KINETICS AND CATALYSIS

Vol. 54

No. 2

2013

CARBON MONOXIDE HYDROGENATION

270°C, but reaches the level of 2.45–2.48 L/h when this temperature is reached on return (already in the CSI mode). This hysteresis cannot be due to an exper imental error. Rather, the hysteresis is a consequence of the formation of other carboncontaining products (Fischer–Tropsch synthesis), a fact that is confirmed by chromatographic analysis data (Fig. 6). The amount of the product hydrocarbons varies on passing to the CSI mode (Fig. 6). (Note that methanol has not been found in the sample taken immediately at the reactor outlet.) Although the product composition does not change qualitatively after catalyst surface ignition, the amount of hydrocarbons dramatically decreases and the concentration of methane (the most abundant peak on the chromatogram) increases. Thus, the yield of hydrocarbons is substantially higher before than after the ignition, since hydrogen and CO after the ignition are largely consumed for the forma tion of methane at the bed entrance. Judging from the ratio between methane and hydrocarbons before and after the transition of the reaction to the CSI regime, they seem to be formed on different sites. Comparing the chromatograms with the available data tabulated for a selected column and the results of identification of individual compounds, we have come to the conclusion that the samples contain С2–С5 hydrocarbons along with methane at elution times no longer than 20 min. The chromatographic peaks corresponding to elution times of 8 to 12 min can belong in the successive order to nbutane, butene1, transbutene2, cisbutene2, and isopentane, and those with elution times of 13 to 19 min can be attributed to npentane, pentene1, transpentene2, cispentene2, etc. Note that hydrocarbons are likely to be produced in insignificant amounts in the thin bed as well, thereby affecting both the ratio between the С1 products and the yield of methane. To summarize, the CO methanation reaction on the 5%Ru/Al2O3 catalyst is easy to transfer to the externaldiffusion regime, the catalyst surface ignition mode, by either heating the catalyst in the reaction medium or replacing Н2 with the reactant gas at tem peratures above the critical ignition temperature. At a given initial gas composition, methanation is accompanied by the complete consumption of reac tant hydrogen and, as such, allows for the conclusion that hydrogen is the critical component, which deter mines the externaldiffusion regime. Simultaneously with methanation, the watergas shift (reaction of CO with the water produced by methanation) proceeds. Fast temperature oscillations in the hotspot of the cat alyst bed are observed in the CSI mode, which are pre sumably due to the deposition and removal of carbon. The comparison of the product compositions at the outlet of the thick (20 mm) and thinbed (3 mm) reactors has allowed for the conclusion that an endot KINETICS AND CATALYSIS

Vol. 54

No. 2

2013

231

hermic reaction (reverse watergas shift) occurs in colder, lower layers of the thick bed. Therefore, the reaction is faster quenched in the thick rather than the thin bed. The methanation is accompanied by the Fischer– Tropsch reaction: a range of hydrocarbons are formed with their quantitative yield dropping when the meth anation reaction switches to the CSI mode. REFERENCES 1. Chorkendorff, I. and Niemantsverdriet, H., Concepts in Modern Catalysis and Kinetics, Weinheim: WileyVCH, 2007. 2. Krämer, M., Duisberg, M., Stöwe, K., and Maier, W.F., J. Catal., 2007, vol. 251, p. 410. 3. Galletti, C., Specchia, S., and Specchia, V., Chem. Eng. J., 2011, vol. 167, p. 616. 4. Kimura, M., Miyao, T., Komori, S., Chen, A., Higash iyama, K., Yamashita, H., and Watanabe, M., Appl. Catal., A, 2010, vol. 379, p. 182. 5. Djinovic, P., Galletti, C., Specchia, S., and Specchia, V., Catal. Today, 2011, vol. 164, p. 282. 6. FujiyamaNovak, J.H., Huang, C.H., Wall, R.L.V., and Carranza, S., J. Chem. Chem. Eng., 2011, vol. 5, p. 49. 7. FrankKamenetskii, D.A., Diffusion and Heat Transfer in Chemical Kinetics, New York: Plenum, 1969. 8. Han, D.H., Park, O.O., and Kim, Y.G., Appl. Catal., A, 1992, vol. 86, p. 71. 9. Bos, A.N.R., Hof, E., Kuper, W., and Westerterp, K.R., Chem. Eng. Sci., 1993, vol. 48, p. 1959. 10. Kirillov, V.A. and Koptyug, I.V., Ind. Eng. Chem. Res., 2005, vol. 44, p. 9727. 11. Kagan, Yu.B., Ponomarenko, A.T., Rozovskii, A.Ya., Loktev, S.M., and Bashkirov, A.N., Neftekhimiya, 1965, vol. 5, p. 82. 12. Kagan, Yu.B., Ponomarenko, A.T., and Rozovskii, A.Ya., Kinet. Katal., 1966, vol. 7, no. 4, p. 679. 13. Rozovskii, A.Ya., Kipnis, M.A., Volnina, E.A., Lin, G.I., and Samokhin, P.V., Kinet. Catal., 2007, vol. 48, no. 5, p. 701. 14. Rozovskii, A.Ya., Kipnis, M.A., Volnina, E.A., Samokhin, P.V., Lin, G.I., and Kukina, M.A., Kinet. Catal., 2009, vol. 50, no. 5, p. 691. 15. Kipnis, M.A. and Volnina, E.A., Kinet. Catal., 2010, vol. 51, no. 2, p. 279. 16. Kipnis, M. and Volnina, E., Appl. Catal., B, 2010, vol. 98, p. 193. 17. Kipnis, M. and Volnina, E., Appl. Catal., B, 2011, vol. 103, p. 39. 18. Khasin, A.A., Gazokhimiya, 2008, no. 2, p. 28. 19. Dry, M.E., Catal. Today, 2002, vol. 71, p. 227. 20. Kowalczyk, Z., Sto –l ecki, K., RarógPilecka, W., Miskiewicz, E., Wilczkowska, E., and Karpinski, Z., Appl. Catal., A, 2008, vol. 342, p. 35. 21. Panagiotopoulou, P., Kondarides, D.I., and Verykios, X.E., Appl. Catal., A, 2008, vol. 344, p. 45.

232

KIPNIS et al.

22. Jiménez, V., Sanchez, P., Panagiotopoulou, P., Val verde, J.L., and Romero, A., Appl. Catal., A, 2010, vol. 390, p. 35. 23. Eckle, S., Anfang, H.G., and Behm, R.J., Appl. Catal., A, 2011, vol. 391, p. 325. 24. Zhang, Z.G. and Xu, G., Catal. Commun., 2007, vol. 8, p. 1953. 25. Nurunnabi, M., Murata, K., Okabe, K., Inaba, M., and Takahara, I., in Advances in Fischer–Tropsch Synthesis, Catalysts, and Catalysis, Davis, B.H. and Occelli, M.L., Eds., Boca Raton, Fla.: CRC, Taylor & Francis Group, 2010. 26. Nurunnabi, M., Murata, K., Okabe, K., Inaba, M., and Takahara, I., Appl. Catal., A, 2008, vol. 340, p. 203. 27. Zhang, Q., Kang, J., and Wang, Y., Chem. Catal. Chem., 2010, vol. 2, p. 1030. 28. Dasgupta, D. and Wiltowski, T., Fuel, 2011, vol. 90, p. 174. 29. Eckle, S., Anfang, H.G., and Behm, R.J., J. Phys. Chem., 2011, vol. 115, p. 1361. 30. Tada, S., Kikuchi, R., Urasaki, K., and Satokawa, S., Appl. Catal., A, 2011, vol. 404, p. 149.

31. Kipnis, M.A., Samokhin, P.V., Bondarenko, G.N., Volnina, E.A., Kostina, Yu.V., Yashina, O.V., Bara banov, V.G., and Kornilov, V.V., Russ. J. Phys. Chem. A, 2011, vol. 85, p. 1322. 32. Borodko, Y. and Somorjai, G.A., Appl. Catal., A, 1999, vol. 186, p. 355. 33. Zhilyaeva, N.A., Volnina, E.A., Shuikina, L.P., and Frolov, V.M., Pet. Chem., 2000, vol. 40, p. 383. 34. Catalyst Handbook with Special Reference to Unit Pro cesses in Ammonia and Hydrogen Manufacture, London: Wolfe Scientific Books, 1970. 35. Freund, H.J., Meijer, G., Scheffler, M., Schlögl, R., and Wolf, M., Angew. Chem. Int. Ed., 2011, vol. 50, p. 10064. 36. Rosenthal, D., Girgsdies, F., Timpe, O., Weinberg, G., and Schlögl, R., Z. Phys. Chem., 2011, vol. 225, p. 57. 37. Wang, M., Weng, W., Zheng, H., Yi, X., Huang, C., and Wan, H., J. Nat. Gas Chem., 2009, vol. 18, p. 300. 38. Tsotsis, T.T., Rao, V.U.S., and Polinski, L.M., AIChE J., 1982, vol. 28, p. 847.

KINETICS AND CATALYSIS

Vol. 54

No. 2

2013