Catalysts

Abstract

The dehydrogenation of an alkane to an alkene, especially isobutane to isobutene, is carried out in admixture with oxygen and in the absence of added steam over a dehydrogenation and oxidation catalyst comprising a platinum group metal deposited upon a support. A catalyst comprising platinum deposited on a support which is a mixture of tin oxide and zirconium oxide possesses good activity for the dehydrogenation of an alkane, especially carried out in this way.

Description

FIG. 3 is a schematic diagram of a reactor apparatus of the present invention.

The invention will now be described with reference to the following Examples.

In the prior art, the Pt and Sn are usually supported on Al₂O₃, with the loading of Sn being ≦5% measured as tin (see J C Hayes, U.S. Pat. No. 4,003,852). Although there are some references to the use of ZrO₂ (E Clippinger and B F Mulaskey, U.S. Pat. No. 3,864,284; G J Antos, U.S. Pat. No. 4,003,826; J C Hayes, U.S. Pat. No. 4,003,852), its function has been claimed to be simply that of a physical support.

EXAMPLE 1 AND COMPARATIVE EXAMPLE 1

Pt-Sn/Al₂O₃ is a known hydrocarbon-conversion catalyst, which is effective for reactions such as reforming (see T-H Chao et al U.S. Pat. No. 5,128,300) and direct dehydrogenation (see J W Jenkins, U.S. Pat. No. 3,511,888) of C₂-C₂₀ alkanes. A catalyst with the nominal composition (by mass) of 1%Pt-1%Sn/Al₂O₃ was prepared (following the method described by F C Wilhelm, U.S. Pat. No. 3,998,900) by co-impregnating γ-Al₂O₃ with an aqueous mixture of hydrogen hexachloroplatinate(IV) (chloroplatinic acid) and acidified tin(II) chloride. The resultant material was dried (110°C C.; air;, 24 hr) and calcined (500°C C.; air; 2 hr). As is conventional, small amounts of tin oxide are measured and written as Sn and larger amounts, eg 10%, are measured and written as SnO₂.

A packed bed (1 cm³) of powdered sample (<150 μm particle diameter) was tested in an adiabatic reactor. For measurements of direct-dehydrogenation activity at 450°C C., in Comparative Example 1, a gas-feed of undiluted isobutane was used at a flow-rate of 50 cm³ min^-1 (GHSV=3000 hr^-1; MHSV=6 dm³ hr^-1g_cat^-1). The molar conversion (% isobutane converted to all products) and selectivity (number of moles of isobutane converted to isobutene divided by number of moles of isobutane converted to all products) were recorded as a function of time at selected furnace/gas-inlet temperatures; the molar yield was calculated from the relationship: $\mathrm{Yield}/\%=\frac{\mathrm{conversion}/\%\times \mathrm{selectivity}/\%}{100}$

Oxidative dehydrogenation was carded carried out, in Example 1, by adding just enough air to the gas-feed to ensure thermally-neutral operation (ie i.e., bed temperature=furnace/gas-inlet temperature). The space velocity of the isobutane was the same, therefore, as during direct dehydrogenation. Again, the molar conversion and selectivity were recorded as a function of time.

In both modes of operation (direct dehydrogenation and oxidative dehydrogenation), the catalyst showed very high selectivity (≧95%) toward isobutene formation. Only during the first 5 minutes of testing was there any sign of an undesired cracking product (propene). In the oxidative mode, the amount of CO₂ formed was just above the detection limit of the GC analyser, no CO was detected.

As shown in Table 1, the direct-dehydrogenation activity declined noticeably during the first 60 minutes; thereafter, the de-activation was very gradual. The initial loss in activity coincided with the bed temperature decreasing to a new stable value, as the endothermic reaction reached steady-state.

TABLE 1

(Comparative Example 1)
Direct Dehydrogenation over 1% Pt-1% Sn/Al2O3
                 Isobutene Yield/%
Elapsed Time/min 450°C C.

2                16.3
25               15.4
55               15.2
120              15.1
300              14.7
1800             --

--: not recorded

When the catalyst was tested in the oxidalive mode, at 450°C C. and 500°C C., the reaction became thermally neutral when the oxygen concentration reached ca 3 or 4% at 450°C C., and ca 5.5% at 500°C C. The initial activity at 450°C C. was higher than for direct dehydrogenation (compare Tables 1 and 2). The isobutene yield is higher at 500°C C. than at 450°C C.

TABLE 2
(Example 1)
Oxidative dehydrogenation over 1% Pt-1% Sn/Al2O3
	Isobutene Yield/%
Elapsed Time/min	450°C C.	500°C C.
2	24.6	27.8
25	18.7	24.6
55	18.0	23.0
100	--	21.6
180	16.4	19.2
300	15.5	16.5
400	14.9	--

EXAMPLE 2

The catalyst described in Example 1 and Comparative Example 1 was used to dehydrogenate isobutane at 450°C C., under the conditions described in Comparative Example 1. The yield of isobutene was allowed to decline to 15%, before air was added to the gas-feed. The activity of the catalyst was then measured as a function of gas-feed composition (Table 3).

TABLE 3
Oxidative dehydrogenation of isobutane over
1% Pt-1% Sn/Al2O3 at 450°C C.
% Air in	Bed Temperature	Isobutane	Isobutene
Gas-Feed	°C C.	Conversion %	Selectivity %
70	498	24.0	70
65	488	21.8	78
55	478	20.0	84
45	468	20.8	85
35	460	19.5	90
20	450	18.0	95

At high air concentrations, the catalyst bed temperature exceeded the furnace temperature and the major products were isobutene and carbon dioxide. As the concentration of air was lowered, the bed temperature decreased and the selectivity to isobutene improved. An optimum gas-composition was eventually achieved, which resulted in the bed temperature remaining at 450°C C., whilst very little carbon dioxide was formed.

EXAMPLE 3

1%Pt-1%Sn/ZrO₂ (nominal composition, by mass) was prepared by the method described in Example 1 and Comparative Example 1, except that zirconia was substituted for γ-Al₂O₃. The catalyst was subjected to the same tests (at isobutane-GHSV=3000 hr^-1) as described in Example 1 and Comparative Example 1, except that both oxidative and direct dehydrogenation were carried out at 500°C C. as well as 450°C C.

TABLE 4
Direct Dehydrogenation Over 1% Pt-1% Sn/ZrO2
	Isobutene Yield/%
Elapsed Time/min	450°C C.	500°C C.
2	15.7	29.2
25	15.2	25.7
60	15.1	25.4
240	14.5	23.6
500	--	22.4
1150	13.4	--
1380	13.4	19.0
1500	13.4	18.6

The marked improvement at 500°C C. can be seen.

In the oxidative mode, there was again a marked improvement at 500°C C. compared to use at 450°C C. At 500°C C., the zirconia-containing catalyst showed improved durability over the 1%Pt-1%Sn/Al₂O₃ catalyst (Tables 2 and 5), with the isobutene yield exceeding 20% during the first 185 min of testing (compared to 150 min for Pt-Sn/Al₂O₃).

TABLE 5
Oxidative Dehydrogenation Over 1% Pt-1% Sn/ZrO2
	Isobutene Yield/%
Elapsed Time/min	450°C C.	500°C C.
2	19.6	31.6
25	18.3	29.0
55	17.8	26.5
120	15.2	23.1
180	14.6	20.0
300	14.0	18.0

EXAMPLE 4 AND COMPARATIVE EXAMPLE 2

1%Pt-1%Sn/10%ZrO₂-Al₂O₃ (nominal composition, by mass) was prepared by the method described in Example 1 and Comparative Example 1, except that ZrO₂-Al₂O₃ was substituted for Al₂O₃. The mixed-oxide was made by impregnating γ-Al₂O₃ with an aqueous solution of zirconium nitrate, before drying (110°C C.; air; 24 hr) and calcining (500°C C.; air; 2 hr).

The good durability gained by using ZrO₂ (in Example 3) was lost by adding Al₂O₃. During oxidative testing (Example 4) at 500°C C. under identical conditions to those of Example 3, the yield of isobutene dropped from 30.0% to 15.0% in the course of the first 85 min. This shows the deleterious effect of alumina; the catalyst does not have the present supporting amount of the mixture of tin oxide and zirconium oxide.

EXAMPLE 5

1%Pt/10%SnO₂-ZrO₂ (nominal composition, by mass) was prepared by co-precipitating SnO₂ and ZrO₂ from an aqueous mixture of tin(IV) chloride and hydrated zirconium oxychloride, using aqueous sodium hydroxide as the precipitant. The precipitate was washed thoroughly, before being dried (110°C C.; air; 24 hr) and calcined (500°C C.; air; 2 hr). The resultant mixed-oxide was impregnated with aqueous dinitrodiammine platinum(II), before the above drying and calcination steps were repeated. The catalyst was tested under the conditions described in Example 3.

The increased loading of Sn was not beneficial to direct dehydrogenation, but resulted in improved durability during oxidative dehydrogenation (compare Tables 5 and 6). At 500°C C., the isobutene yield exceeded 20% during the first 210 minutes (compared to 185 minutes for 1%Pt-1%Sn/ZrO₂).

TABLE 6
Oxidative Dehydrogenation Over 1% Pt/10% SnO2--ZrO2
	Isobutene Yield/%
Elapsed Time/min	450°C C.	500°C C.
2	17.1	28.2
25	18.4	28.7
55	18.2	27.6
120	16.5	24.2
180	14.8	21.3
240	--	18.7

EXAMPLE 6

1%Pt/20%SnO₂-ZrO₂ (nominal composition, by mass) was prepared by the method described in Example 5, and tested under the conditions described in Example 3. Again, the clearest benefit derived from the high tin loading was apparent in the oxidative mode, both at 450°C C. and 500°C C., when the rate of deactivation was even further reduced (compare Tables 5, 6 and 7). In particular, the durability at 450°C C. (as measured by the duration of yield ≧15%) exceeded 24 hr (compared to 6 hr for 1% Pt-1%Sn/Al₂O₃); see FIG. 1.

TABLE 7
Oxidative dehydrogenation over 1% Pt/20% SnO2--ZrO2
	Isobutene Yield/%
Elapsed Time/min	450°C C.	500°C C.
2	19.0	38.7
25	18.8	31.4
55	18.6	29.0
180	18.0	24.0
300	17.6	20.8
400	--	18.8
1260	15.1	--
1500	14.9	--

EXAMPLE 7

A fresh sample of 1%Pt/20%SnO₂-ZrO₂ (as described in Example 6) was treated under oxidative conditions, but at half the space velocity used in Examples 1-6 and Comparative Examples 1 and 2.

At this lower space velocity (isobutane-GHSV=1500 hr^-1), the catalyst de-activated more gradually. As FIG. 2 shows, its initial activity at 500°C C. was similar to that observed in Example 6, but the yield still exceeded 25% after 5 hours (the time taken for the yield to fall below 20% at isobutane-GHSV =3000 hr^-1).

COMPARATIVE EXAMPLE 3

1%Pt/SnO₂ (nominal composition, by mass) was prepared by impregnating SnO₂ with an aqueous solution of tetraammineplatinum(II) hydroxide, before drying (110°C C.; air; 24 hr) and calcining (500°C C.; air; 2 hr). It was tested under the conditions described in Example 3. This material was a very poor catalyst for both direct dehydrogenation (2.0% maximum yield at 450°C C.) and oxidative dehydrogenation (3.2% maximum yield at 450°C C.).

EXAMPLE 8

The ability of 1%Pt/20%SnO₂-ZrO₂ (as described in Example 6) to dehydrogenate linear alkanes was tested by following the procedures given in Example 3, but replacing the isobutane reactant with normal butane. During direct dehydrogenation, the initial total yield of unsaturated products was 26% (product selectivity: 32% 1-butene, 27% cis 2-butene, 38% trans 2-butene, 2% butadiene), but declined to 14% in the space of 3 hours. On switching to the oxidative mode, without first regenerating the catalyst in any way, the total yield recovered, reaching a maximum of 29% before declining slowly (down to 25% after a further 2 hours); the product distribution was very similar to that observed during direct dehydrogenation.

EXAMPLE 9

The sequence of tests described in Example 8 was repeated using propane as the alkane reactant. During direct dehydrogenation, the initial yield of propene was 19%; after 3 hours, it had declined to 12% On switching to the oxidative mode, the yield was restored to 19%. Thereafter, it declined slowly to 17% during the next 4 hours.

The invention is further illustrated by the yield data given graphically for various catalysts in the accompanying FIGS. 1 and 2.

FIG. 1:

Oxidative dehydrogenation of isobutane (GHSV=3000 hr^-1) at 450°C C., over

(a) 1%Pt/20%SnO₂-ZrO₂;

(b) 1%Pt-1%Sn/Al₂O₃.

FIG. 2:

Oxidative dehydrogenation of isobutane at 500°C C., over

(a) 1%Pt/20%SnO₂-ZrO₂;

(b) 1%Pt/20%SnO₂-ZrO₂;

(c) 1%Pt-1%Sn/ZrO₂;

(d) 1%Pt-1%Sn/Al₂O₃.

For (a), GHSV=1500 hr^-1; (b)-(d); GHSV=3000 hr^-1

Claims

1. A catalyst for alkane dehydrogenation, comprising by weight 0.1 to 3% platinum, calculated as metal, 6 to 60% tin oxide, and 37 to 94.9% zirconium oxide, the platinum deposited upon a support which is a mixture of the tin oxide and the zirconium oxide.

2. A catalyst according to claim 1, containing 10 to 60% by weight of the tin oxide.

3. A catalyst according to claim 1, wherein the support contains substantially no alumina.

4. A catalyst according to claim 1, consisting essentially of 0.1 to 3% by weight of platinum, calculated as metal, 10 to 60% by weight of tin oxide, 37 to 94.9% zirconium oxide, and at least one member selected from the group consisting of stabilisers and promoters.

5. A catalyst according to claim 1, wherein the support comprises SnO₂ and ZrO₂ in a weight ratio of approximately 1:4.

6. A catalyst according to claim 5, comprising approximately 1% by weight of platinum, impregnated onto a co-precipitate of SnO₂ and ZrO₂ in a weight ratio of approximately 1:4.

7. A catalyst according to claim 1, wherein the catalyst has substantially an absence of alumina.

8. A catalyst according to claim 1, wherein the catalyst contains 15-30% by weight of tin oxide.

9. A catalyst according to claim 1, wherein the catalyst contains 70-85% by weight zirconium oxide.

10. A catalyst according to claim 1, wherein the catalyst has a weight ratio of the tin oxide to the zirconium oxide of 1:3-9.

11. A catalyst according to claim 1, wherein the catalyst has a weight ratio of the tin oxide to the zirconium oxide of 1:3-5.

12. A catalyst for alkane dehydrogenation comprising a catalytically effective amount of platinum, 6 to 60% tin oxide, and 37