Al2O3 catalyst

Al2O3 catalyst

NATURAL GAS CONVERSIONV Studies in Surface Science and Catalysis, Vol. 119 A. Parmaliana et al. (Editors) 1998 Elsevier Science B.V. 659 O x i d a t...

263KB Sizes 0 Downloads 16 Views

NATURAL GAS CONVERSIONV Studies in Surface Science and Catalysis, Vol. 119 A. Parmaliana et al. (Editors) 1998 Elsevier Science B.V.

659

O x i d a t i v e d e h y d r o g e n a t i o n o f p r o p a n e in a n n u l a r reactor o v e r a Pt/A1203 catalyst A. Beretta a, M. E. Gasperini a, G. Trepiedi a, L. Piovesan b, P. Forzatti a aDipartimento di Chimica Industriale e Ingegneria Chimica "G. Natta", Politecnico di Milano, piazza L. da Vinci, 32, 20133 Milano, Italy bSnamprogetti/Rice, via Maritano 26, 20097 San Donato Milanese (MI), Italy

Oxidative dehydrogenation of propane was tested inside a structured annular reactor with and without a Pt/A1203 catalyst. The Pt-containing catalyst was very active in the total combustion of propane, which was the prevailing reaction at temperatures lower than 500~ At higher temperatures the formation of olefins was observed; this was very likely assisted by gas-phase reactions. Apparently, the presence of the catalyst shifted of 50~ towards lower T the onset of the ODH reaction with respect to the single homogeneous process.

1. INTRODUCTION Among the novel routes for the chemical conversion of natural gas, oxidative dehydrogenation (ODH) of light alkanes represents a potential alternative to the traditional endothermic processes for the production of short-chain olefins. A recent review of the best results obtained in the oxidative dehydrogenation of ethane and propane has been provided by Baerns and Buyevskava [1]. They pointed out that, although a number of catalytic systems have been proved to be active in the ODH reactions including V-Mg-O systems, phosphates and molybdates of various transition metals, and structured noble metal-based systems, the development of a catalyst with economically feasible performance is still a "challenging task". Very interesting data were obtained in the ODH of ethane and propane by Schmidt and coworkers; they observed almost complete conversion of the alkanes and selectivity as high as 60-70% of the olefins by using a Pt-coated foam monolith in an autothermal reactor at very short contact times [2]. However, due to the very high reaction temperature associated to the adiabatic reactor configuration (900-1000~ possible contributions from homogeneous reactions were likely present; the effective role of the catalyst in the process seems thus unclear. The present work addresses the study of the ODH of propane over a Pt/A1203 catalyst. Experiments were performed by using a novel annular reactor, wherein the catalyst is deposited as a thin and short layer onto a tubular ceramic support co-axially inserted inside a quartz tube. The gas stream flows across the annular chamber; given the absence of pressure drop, high flow rates can be guaranteed and, as very small amounts of catalyst can be

660 deposited (10-100 mg), gas hourly space velocities as high as 106 L(STP)/kgcat/h are easily realized [3, 4]. Also, the peculiar geometry of the system favors the dispersion of the heat produced by the reaction via radiation towards the oven internal wall; even in the presence of very exothermic processes, thus, a relatively good control of the catalyst temperature can be obtained and almost isothermal conditions arise along the catalyst bed. The catalyst temperature, then, can be read by exploiting the tubular support as a thermocouple-well. Experiments were performed both in the presence and in the absence of catalyst in order to better understand the contribution of reactivity associated to gas phase reactions. 2. OXIDATIVE DEHYDROGENATION OF PROPANE OVER Pt/AI203 Few milligrams of a commercial 3% Pt/A1203 catalyst were deposited onto the ceramic support and a 13 mm long and about 50 ~tm thick catalyst layer was obtained. The activity tests were performed at a total flow rate of 120 Ncc/min, with feed composition C3H8 : 02 : N2 = 1 : 1 : 4, at atmospheric pressure and by increasing the temperature up to 700~ The flow rate to catalyst surface ratio herein used corresponds to the same value realized by Huff and Schmidt [2] in the foam monolith reactor at 9 STPL/min feed stream. In other words, the same "contact time" referred to the catalyst geometric surface was reproduced as a reference. Referred to the catalyst load, the flow rate corresponded to a gas hourly space velocity of 1.2 106 L(STP)/kg cat/h. At a oven temperature as low as 150~ the catalyst was already very active in the complete combustion of propane to CO2 and water. The reaction rate was so high that the process was strongly limited by inter-phase diffusion resistances; as shown in Figure 1 in the range T>400~ propane and oxygen conversions were in fact almost constant up to 500-550~ 100

r

-

.

02

80 E

.o (/) L_

G) > E 0 (O

6o

40

C3H8~ 20

0

400

,

I

450

,

I

500

,

I

550

,

I

,

600

Oven Temperature,

I

650

~

I

700

~

Figure 1 - Propane and oxygen conversions at increasing oven temperature over Pt/A1203

661 Beyond such temperature, propane conversion increased rapidly and was almost complete at 750~ Oxygen conversion was always high in the whole range of temperature explored, so that the overall stoichiometry of the process changed from C3H8 : O 2 = 1 : 5 (stoichiometry of the combustion reaction) to C3H8 : 02 - 1 : 1. While at lower temperature CO2 and water were the only reaction products, the increase of propane conversion was accompanied by a change in the product distribution. This is shown in Figure 2, where the % C-mole and % H-mole selectivities of the reaction products are plotted as functions of the oven temperature. 100

),, > tO tO

8O

60 03H6

C2H 4

40 T

d

Q

~

2o

o

550

4 2

575

600

625

650

675

700

Oven Temperature ~ 100

8O

H20

.....

> .,.,_

Hydrocarbons

"~ (1) 60 (l) 03 0

40

E I

"1-

~

20 / e 0

550

H2 ,

575

600

I

625

,

I

650

i

I

675

,

700

Oven Temperature ~ Figure 2 - Distribution of the C-containing and H-containing reaction products.

662 The presence of propylene traces in the product mixture was first observed at about 450~ oven temperature. The amount of the olefin became significant at 550~ over this temperature also ethylene was observed, and the formation of both olefin increased remarkably. The selectivity of propylene increased up to a maximum (28-30%) at Toven 625~ At higher temperatures its production decreased, probably due to the increasing relevance of secondary reactions, responsible for the formation of shorter C-containg species. The selectivity of ethylene and methane increased progressively. At high temperatures, while the formation of CO2 and water had a dramatic drop, CO and H2 concentrations in the product mixture increased. The highest total yield to olefin was obtained at 640~ with propylene + ethylene selectivity of 60% and propane conversion of 78%. As mentioned in the Introduction, the mullite tube (coated with catalyst in a short central portion) was used as thermocouple-well, so that during each experiment the axial temperature profiles were measured. The axial temperature profile of the oven wall was also measured. While upstream and downstream from the catalytic portion mullite temperature and oven temperature were almost the same, the catalyst temperature was generally higher than the heating temperature. Figure 3 shows the relationship between oven and catalyst temperatures in the various experiments. At lower reaction T, in correspondence with the prevailing highly exothermic combustion route, the catalyst stabilized at nearly 150~ over the oven temperature. This gap decreased progressively at increasing oven T along with the lowering of CO2 selectivity and the onset of the much less exothermic oxidative dehydrogenation.

700 w

650

0 0

_

600

S

I,-,

550 E}.. E 500

I450

400 400

I

I 450

I

I 500

t

I 550

I

I

I

600

Oven Temperature,

I 650

J

I 700

~

Figure 3 - Catalyst temperature vs. oven temperature during activity tests.

663

3. GAS-PHASE OXIDATIVE D E H Y D R O G E N A T I O N OF PROPANE

Figures 4 and 5 show the results of the experiments performed in the absence of catalyst under the same operating conditions of the catalytic tests. It was found that the oxidative dehydrogenation of propane was thermally activated at nearly 600~ and a 90% conversion of the paraffin was obtained at the heating temperature of 710~ Oxygen conversion was always slightly lower than propane conversion. 100

,

,

,

,

,

,

,

,

,

,

,

I

'

80

Propane ~.....~

r .9

60

I...,

(D > tO

40

0 20

0

B --

'

I

575

--

'

I

600

'

I

625

'

I

650

'

675

7'00

725

Oven Temperature, ~

Figure 4 - Thermal activation of the C3Hs/O2/N2=l/1/4 mixture. Conditions as in Fig.1 Propylene and ethylene were the most abundant products at lower propane conversions. At increasing temperature, propylene selectivity decreased while the selectivity to terminal products like CO and CH4 increased. CO2 was always present in very small amounts. 60

' ..... ~

I

'

I

'

I

I

'

I

'

I

'

I

'

I

'

I

"'

I

50

.# 9. ~ O

40

E

2o

CO~--o~-o . . . . 0

'

640

I

650

'

660

670

680

'

I

'

690

I

700

o 710

'

720

Oven Temperature, ~ Figure 5 - Distribution of C-containing products of homogeneous propane ODH

664 Catalytic and homogeneous data are compared in Figure 6, where the outlet flow of propylene and ethylene are plotted vs. the reactor temperature. Both in the presence and in the absence of catalyst, the same reaction mechanism seemed to govern the formation of olefins; within this mechanism propylene behaved as intermediate product (going through a maximum productivity), while ethylene presented the feature of a terminal product. However, the presence of the small amount of catalyst (a negligible volume in comparison with the total gas-phase volume in the annular reactor) "anticipated" of about 50~ the activation of the propane/oxygen mixture in the production of olefins. I

'

I

~

I

'

I

~

I

'

I

'

I

'

5. Ox 10 .4 ethy[ene

t'.,~.

(~a~~

~

y

4.0x10-4

E (!) 0

3.0x10. 4

E (l)

propylen

2.0X10"4

~

t._

~:

1.0xlO -4

o

. .0. .

propyene( omi

LL _=

0.0

'

475

_=

I

500

'

I

I

525

550

'

I

575

'

I

600

'

I

625

'

I

650

'

I

675

'

I

700

'

725

Oven Temperature, ~ Figure 6 - Comparison of ethylene and propylene flow rates with and without catalyst. 4. CONCLUSIONS The present data have clarified that: (1) the Pt-supported catalyst is active in the total oxidation of propane; this reaction prevails at temperatures lower than 500~ maybe due to strongly adsorbed O-species which are very aggressive towards the paraffin; (2) the thermal activation of propane and oxygen mixtures largely contributed to the overall process at high temperatures (>625-650~ The effective contribution of the catalyst to the formation of olefins is still open. The presence of catalyst lowered the T-threshold for the production of propylene; further research is needed to better understand whether this effect resulted from catalytic and/or thermal factors. REFERENCES 1. M. 2. M. 3. A. 4. A.

Baems, O. Buyevskava, Proceedings of JECAT'97, p. 21, 1997. Huff, L. D. Schmidt, J. Catal. 149 (1994) 127. Beretta, Chem. Eng. Comm., (1998) in press. Beretta, P. Baiardi, D. Prina, P. Forzatti, Chem. Engng. Sci, (1997)submitted