OXIDATIVE DEHYDROGENATION PROCESS AND CATALYST
CROSS REFERENCE TO RELATED APPLICATIONS
This application is claims benefit to provisional application 60/117,095 filed January 25, 1999, which is incorporated by reference in its entirety for all useful purposes.
FIELD OF THE INVENTION
10 This invention pertains to the production of olefins by the oxidative dθhydrogenation or cracking of saturated hydrocarbons over a catalyst containing chromium and/or chromium oxide. High yields of olefins can be obtained over structured catalysts with very short contact times and no catalyst deactivation.
BACKGROUND OF INVENTION
Currently, olefins are used as important chemical intermediates for a large number of industrial processes and consumer products. These light olefins such as ethylenβ, propylene, and butenes are usually produced by the steam pyrolysis process [Pyrolysis: Theory and Industrial Practice; Albright, L. F, et al, Eds.; Academic Press: New York, 1983]. The two major limitations of the current steam pyrolysis process are the high gas temperatures (Θ00-900°C) required and the deposition of coke on the walls of the reactor tube that require the periodic shut-down of the process for decoking. To overcome these limitations, there has been much recent interest in trying to find a catalytic alternative to the current industrial steam cracking process.
_;> Many researchers have investigated the catalytic dβhydrogenation of light alkanes as a route to obtaining alkenes for polymerization and other organic synthesis. This approach does not cofββd an oxidant. Much of this research has focused on
E.. J. Catai. 1993, 142, 166-171 ; Hoang, M.; Hughes, A. E. et al, J. Catal. 1997, 171, 313-319; Hoang, M. et al, React. Kinet. Catal. Lett. 1997, 61, 21-26; Zaki, M. I. et al, Appl, Catal. 1986, 21, 359-377], Thus, the catalytic properties of supported chromium catalysts are strongly affected by the acidity/basicity of the oxide support. However, there is significant disagreement over the exact oxidation state of the active Cr surface species. Recent research has also examined the use of supported chromium oxide catalysts for the catalytic dehydrogenation of hydrocarbons [De Rossi, S. et al, Appl. Catal. A: General 1992, 81, 113-132; De Rossi, S. et al, Appl. Catai A: General 1993, 106, 125-141; De Rossi, S. et al, J. Catal, 1994, 148, 36-46; Lugo, H. J. and Lunsford, J. H., J. Catal. 1985, 91, 155-166; Udomsak, S. and Anthony, R. G., Ind. Eng. Chem Res. 1996, 35, 47-53].
SUMMARY OF THE INVENTION
This invention describes the production of olefins by catalytic partial oxidation of saturated hydrocarbons on chromium oxide supported on monoliths at very short contact times. This catalyst and process can be significantly less expensive to operate than either a standard packed bed reactor using a selective oxide catalyst at a residence time in excess of 0.1 seconds or the newer short contact time oxidative dehydrogenation process using monolith ically supported noble metals. Additionally, this catalyst produces less by-product GO which is advantageous to downstream processes. The monoliths may be composed of any oxide stable at temperature in excess of 900OC including α-A|203, SiC , Mg-stabilized Zrθ2, and Y-stabilized Zrθ2 and may have a honeycomb or foam structure although the foam structure is preferred.
The support affects the activity of the chromium oxide catalyst as does the presence of various promoters, including Cu, Mn, Mg, Ni, Fe, their oxides, and combinations
thereof. The preferred promoters are copper and copper oxides. The preferred metal atom ratio of promoter to chromium is between 1 :100 and 1 :2 and more preferably between 1 :15 and 1 :5.
To summarize these results, supported chromium oxide has been shown to be an effective catalyst for the oxidative dehydrogenation of ethane. As reported in the literature, the activity of the supported chromium oxide catalyst has been shown to depend upon the ceramic support material [Hoang, M. et al, J. Catal. 1997, 171,
313-319], the surface area, and the chromium oxide loading (with a lower loading showing a lower activity and shorter lifetime). The high loading chromium oxide catalyst supported on ZrO≥ monolith showed the best overall activity of all the supported chromium oxide catalysts, but could not achieve stable operation at flow rates higher than 5 SLPM. The 9 wt. % Cr2θ3/Zrθ2 catalyst outperformed the Pt-coated monolith with a higher C2H4 selectivity and C2Hg conversion, and thus higher C2H4 yield. It was also found that the addition of copper to the chromium oxide on Zrθ2 catalyst significantly increased the activity of the catalyst such that stable operation with no signs of deactivation was with GHSV (STP) up to 1 x 1 θβ hr"1. The addition of copper to the Cr2θ3/Zrθ2 catalyst, however, did not significantly affect the product selectivity or reactant conversions.
DETAILED DESCRIPTION OF INVENTION
Any short contact time reactor can be used. The reactor used here is essentially identical to that previously described for the production of syngas [Hickman, D. A. and Schmidt, L. D., J. Catal. 1992, 136, 300-308 and references therein] and oxidative dehydrogenation of light alkanes [ Flick, D. W. and Huff, M. C, J. Catal. 1996, 178, 315-327 and references therein; Huff, M. and Schmidt, L. D., J. Phys. Chem. 1993, 97,
1 1815-11822 and references therein]. The reactor consists of a quartz tube with an inner diameter of 20 mm. The catalyst is sealed in the tube with high temperature silica-alumina felt which prevents the bypass of gases around the catalyst. To reduce the radiation heat loss in the axial and radial directions and to better approximate adiabatic operation, inert foam monoliths are placed in front and behind the catalyst as heat shields, and the reaction zone is externally insulated.
A ceramic foam monolith with 20-100 pores per linear inch (ppi) is impregnated with a Cr containing precursor such as an aqueous solution of Cr(N03)3, and then calcined in air at 600°C for at least four hours. After calcination, the chromium oxide coated monoliths were bright green in color. The chromium oxide can be supported on monoliths made from high temperature (>900°C) stable oxides including α-Al2θ3, Si02l Mg-stabilized Z1O2, and Y-stabiiized Zrθ2- Pt and mixtures of promoters (Cu, Mn, Mg, Ni, Fe, and combinations thereof) were deposited on the monoliths similarly using H2PtCle or transition metal nitrate salts, respectively, in an aqueous solution. The monoliths used measured 10 mm thick and between 18-20 mm in diameter.
The gas flow for the reactor is controlled by electronic mass flow controllers. The total feed flow rate to the reactor ranged from 0.1 to 12 SLPM which corresponds to approximate contact times of 2-200 milliseconds (GHSV = 1 x 105 - 2.2 x 106 hr"1 at STP) for the monolith catalyst. The reactor operates near atmospheric pressure. The reaction occurs autothermalty around 900°C.
While the reaction operates autothermalty at steady state, an external heat source is necessary to ignite the reaction. The gas mixture wfth C2Hg:θ2:N ■
25:15:60) ignites at approximately 350°C over the chromium oxide catalyst at a flow rate of 3 SLPM which is higher than the 220°C preheat needed to ignite that reaction over a Pt coated monolith catalyst. After ignition, the external heat source is removed or
reduced to the desired level of heat addition, the composition is adjusted to the desired value and steady state is established, After each change in feed conditions, the reactor is allowed to achieve steady state (<10 min.) before analysis of the reaction products by the GC. The feed gas consists of C H6 and O2 with N2 as the diluent. The level of dilution ranged from <1 to about 80%. The N2 was used as an internal GC calibration standard. Additional trials have been completed using alternative diluents: N2. He, Ar,
CO2, H2O and mixtures thereof. Some experiments were conducted with the feed composition within the flammability limits. In these instances, it is important to realize that this is a flow system with linear velocities in excess of the flame speed such that homogeneous ignition is impossible [Bolk, J. W. et al, Chem. Eng. Sci. 1996, 51].
The reaction temperature was measured by a type K (chromel/alumel) thermocouple inserted from the rear of the reactor and placed at the center of the reactor tube between the catalyst and the rear radiation heat shield. The temperature measured at the back of the catalyst, the reaction temperature, is a good measure of both the product gas phase temperature and the surface temperature of the rear face of the catalyst, since the gas phase and surface temperatures are approaching equilibrium at the catalyst exit.
In some experiments, the feed gas TO the reactor contained C2H6, O , and H with a small amount of N2 for use as an internal calibration standard. The level of H in the feed ranged from 0 to 50% which corresponds to H O2 ratios ranging from 0 to
3.0. The amount of N2 in the feed was always less than 1%. The process could recycle the H2 and not require a net import of H2 to the system.
The catalyst can contain substantially no Pt. Substantially no Pt means less than 0.1 % Pt by weight, preferably less than 0.01 % Pt by weight.
Table 7 : Carbon atom and hydrogen atom selectivity, Table 8: Carbon atom and hydrogen atom selectivity, conversion, and temperature for ethane oxidation over conversion, and temperature for ethane oxidation over a 45 ppi, 4 t. % CrjOj/ ZrQj monolith as a function a 45 ppi, 6.7% Cu - 6.7% CΓIOJ/ZΓO. monolith as a of flow rate at a C2H«tθ2 ratio of 1.5 with 20% N2 function of flow rate at a CzIfyO. ratio of 2.4 with dilution in an autothermal reactor at a pressure of 1.2 20% Nj dilunon in an autothermal reactor at a pressure arm. of 1.2 aim.
could not be sustained for even a few minutes. Thus, for the Cr2θ3 AI2θ3 catalyst, the reaction can only be sustained inside the flammability range for the C2H6, 02, and N2 system, and the lifetime decreases the closer the mixture is to the upper flammability boundary. However, in stark contrast to the Cr20;_/Al2θ3 catalyst, the 9 wt. % Cr2θ3 Zrθ2 monolith sustained steady reaction up to a C2Hβ 02 ratio of 1.8 with 50% N2 dilution for over 2 hours and at a C2H6/O2 ratio of 1.5 with 50% N2 dilution at 2 SLPM, the Cu modified CraOyZrOs monolith catalyst did not show any deactivation for over 7 hours of continuous operation, while all of the previous chromium oxide containing catalyst rapidly extinguished at a C2H602 ratio of 1.5 with 50% N2 dilution.
The effects of preheat are shown in Tables 9-10 for the 9 wt. % Cr2θ3 α-AI203 monolith catalyst at a C2rV02 ratio of 1.5 with 20% N2 dilution at a flowrate of 2 SLPM. Results are shown as a function of time for nearly adiabatic operation (Table 9) and for
autothermal operation (Table 10) where the reactants have been heated to 250°C before reaching the reaction zone. Preheat results in a catalyst lifetime of -7 hours before the reaction extinguished. The addition of preheat, however, leads to a slightly lower selectivity to C2H4 and a higher selectivity to CO. Preheating the feed also increased the selectivity to CH4 and C2H2 and decreased the selectivity to C02. As expected, the C2H6 conversion increases with preheat.
The most interesting impact of preheat is shown in the temperature of the catalysis. The front and back temperature of the catalyst without preheat and the front temperature of the catalyst with preheat fall dramatically after approximately 1.5 hours of operation. However, the addition of preheat seems to provide enough energy to keep the reaction ignited with relatively stable front and back temperatures over the catalyst for several hours after the catalyst without preheat extinguished. The 250°C preheat of the feed, however, did not provide enough energy to keep the reaction from eventually extinguishing. There were also slight selectivity changes to C2H4, CH4, and CO that correspond with the rapid drop in the temperature at the front of the catalyst.
Table 9: Carbon atom and hydrogen atom selecuviiy, conversion, and temperature for ethane oxidation over a 45 ppi, 9 wt. % CrjCty ct-Al203 monolith as a function of time at a -tyQ. ratio of 1.5 with 20% N2 dilution m an autothermal reactor at a pressure of 1.2 ami. Data is shown for reactants entering the reactor
Table 10: Carbon atom and hydrogen atom selecϋvity. conversion, and temperature for ethane oxidation over a 45 ppi, 9 wt. % CΓICH/ C.ΑI2O5 monolith as a function of time at a C2-tyQ2 ratio of 1.5 with 20% Nj dilution in an autothermal reactor at a pressure of 1.2 atm. Data is shown for reactants entering the reaαor at 250°C temperature. The reaction extinguished between hours 7 and 8.
Time on stream (hr) 0.0 0.5 1.0 1.5 2.0 4.0 6.0 7.0
Carbon atom selectivity (%)
CjH 36.3 36.6 36.5 39.1 40.5 40.4 41.1 37.0
CO 33.2 34.0 34.6 30.0 29.9 30.7 29.5 37.2
C 5.0 4,7 4.3 3.5 3.3 3.3 2.9 3.5
CH4 13.2 13.0 13.0 14.0 14.4 14.3 14.4 12.8
C:H2 9, 1 8.4 7.8 7.9 7.7 7.3 6.6 5.8
Other 3.2 3.3 3.8 5.5 4.2 4.0 5.5 3.7
Hydrogen atom selectivity (%)
H2 46.8 47.2 47.6 42.9 42.8 42,9 40.7 50.8
H,0 14.6 14.4 14.1 14.1 14.4 L4.5 15.1 10.8
Conversion (%)
C2Hβ 98.5 98.3 98.2 98.0 97.9 97.9 97.5 97.8
O, 99.9 99.9 99.9 99.9 99.9 99.9 99.9 99.9
Temperature ( C)
Front 1180 1132 1112 670 640 595 580 565
Rear 1000 962 895 870 840 825 818 760
Example 6: The oxidative dehydrogenation of ethane was investigated with different diluents including N2, Ar, C02 and H20. The carbon and hydrogen atom selectivitiβs, conversions and temperatures for the oxidative dehydrogenation over a 1.0 wt. % Cu - 9.0 wt. % Cr2θ3/Zr02 catalyst at C2rV02 ratios of 1.8 and 2.4 at a total flow rate of 2 SLPM with 20% dilution are shown in Tables 11 and 12. The product selectMties are not significantly affected by changes in the type of diluent. The ethane conversion,
Table 11. Carbon and hydrogen atom Table 12. Carbon and hydrogen atom selectivity, conversion, and selectivity, conversion, and temperature for ethane oxidative temperature for ethane oxidative dehydrogenation over a 1.0 wt. % Cu - dehydrogenation over a 1.0 wt. % Cu - 9.0 wt. % Cr2O3 ZrO2 catalyst at a 9.0 wt. % Cr2O3/ZrO2 catalyst at a C2H O2 ratio of 1.8 with 20% total C2ItyO2 ratio of 2.4 with 20% total dilution at a total flow rate of 2 SLPM dilution at a total flow rate of 2 SLPM in an autothermal reactor at a pressure in an autothermal reactor at a pressure
however, does change slightly with different diluents. Along with the decrease in ethane conversion, the reaction temperature at the back of the catalyst is also lower with the C0
2 and H
20 diluent. Therefore, use of diluents with different heat capacities has the effect of lowering the ethane conversion and temperature at the back of the catalyst with increases in the heat capacity of the diluent used. The use of C0
2 and H
20 which are reaction products in the oxidative dehydrogenation of ethane as the diluent does not significantly affect the product selectivity under these reaction conditions.
Example 7: Table 13 shows typical reactor performance for the oxidative dehydrogenation of propane over a variety of monolithic catalysts at nitrogen dilutions levels of 20% and 50% at total flow rate of 2 SLPM. The less expensive chromium oxide catalyst shows similar selectivity and conversion trends as the Pt catalyst with significantly less production of CO.
Table 13: Carbon atom and hydrogen atom selectivity, conversion, and temperature for propane oxidation over various catalysts at several C3Hg/Oj ratios with 20% and 50% N2 dilution and a total flow rate of 2 SLPM in an autothermal reactor at a pressure of 1.2 aim.
Catalyst 6 wt. % Cr20,/ZrO_ 6 wt. % Pt/α-Al203
Ni dilution (%) 2 0 50 20 50
CJΓ OJ 1.2 1.5 1.8 2.0 1.2 1.5 1.8 1.2 1.5 1.8 2.0 1.2 1.5 1.8 2.0
Carbon atom selectivity (%)
C3H« 1.2 10.3 20.7 26.0 1.1 11.0 22.7 2.8 160 23.5 27.7 1.1 15.1 22.6 26.1 jH* 37.8 45.5 40.7 37.7 34.6 43. L 34.3 37.7 38.7 33.4 32.1 31.3 37.3 29.2 23.8
CO 21.0 9.1 6.3 5.3 23.3 11.0 9.2 26.6 17.0 16.5 13.9 33.6 17.8 18.5 20.7 co2 7.5 8.2 9.0 9.9 9.1 10.9 15.5 3.8 4.1 6.1 6.8 5.8 7.9 12.2 15.0
CH, 23.3 19.8 L6.8 15.6 21.7 17.0 13.0 22.6 18.4 15.7 15.1 20.7 16.7 13.4 1 1.2
C.K . 1.1 2.6 2.5 2.3 0.7 2.4 2.1 1.6 2.4 2.1 2.0 0.9 2.2 1.9 1.5
CjH. 5.9 1.5 0.8 0.5 74 2.0 0.5 3.1 0.8 0.1 0.1 4.5 0.5 0.2 0.1
Other 2.1 2.9 3.1 2.8 2.3 2.6 2.7 1.8 2.6 2.6 2.2 2.1 2.6 2.1 1.6
Eivdrogen atom selecti vity (%)
H2 20.1 9.5 7.4 6.5 19.4 10.0 7.9 16.4 8.2 6.9 5.6 19.9 8.6 8.2 6.8
H,0 13.4 15.9 16.9 17.9 18.9 19.3 26.1 17.5 19.5 24.7 24.6 21.5 23.0 29.5 36.3
Conversion (%)
C3H, 99.9 923 74.3 60.8 100 91.5 64.0 99.3 83.3 63.8 52.5 j 99.9 83.1 57.1 39.3
O, 100 99.5 99.1 99.1 300 99.7 98.9 99.3 98.2 97.7 97.8 1 9.4 98.4 98.0 97.7
Temperature CO
Front 1066 678 584 529 1027 623 500 980 622 545 505 1 1018 941 921 849
Rear 965 875 829 808 934 852 800 969 909 876 866 1 937 831 825 815
surveying a wide variety of metal oxides for high selectivity and conversion [Carra, S. and Fomi, L, Catal. Rev. 1971 , 5, 159-198; Cimino, A. et al, J. Mol. Catal. 1989, 55, 23-33; US Patent #4535067; US Patent #4804799; US Patent #3965044; US Patent #5354935; US Patent #5254788; US Patent #5378350; US Patent #3719721; US Patent #4032589], However, these catalysts generally deactivate with time on stream due to coke build-up on the catalyst surface [Royo, G. et al, Ind. Eng. Chem. Res. 1994, 33, 2563-2570].
Catalytic oxidative dehydrogenation of hydrocarbons is an alternative to thermal pyrolysis and catalytic dehydrogenation. The oxidative dehydrogenation of C-2-C6 hydrocarbons has been examined over noble metal catalysts supported on ceramic foam monolfths [ Flick, D. W. and Huff, M. C, Appl. Catal. A 1999 187, 13-24; Flick, D.
W. and Huff, . C, Abstracts of Papers of ACS 217:018-CATL, Part 2 March 21, 1999;
Flick, D. W. and Huff, . C, Catai Lett 1997, 47, 91-97; Flick, D. W. and Huff, M. C, J.
Catal. 1998, 178, 315-327; Dietz III, A. G. et ai, J. Catal. 1998, 776, 459-473; Huff, M.
C. et ai, Catal. Today 1994, 21, 113-128; US Patent # 5905180 which are all incorporated by reference in its entirety for all useful purposes]. In general, noble metal catalysts are more expensive than oxide catalysts. The oxidative dehydrogenation of ethane over Pt-coated monolith catalysts at short contact times has been investigated concerning the maximization of C2H4 production [US Patent #5382741 ; US Patent
#5625111; Huff, M. and Schmidt, L. D., J. Phys. Chem. 1993, 97, 11815-11822; US Patent #4844837; US Patent #4940826; US Patent #5105052; US Patent #5593935; Wftt, P. M. and Schmidt, L. D., J. Catal. 1996, 163, 465-475; US Patent # 5905180]. These monolith reactors operate at very short contact times on the order of milliseconds. The yield of C2H4 with these reactors has been shown to be about 50% (C2H4 selectivity ~65% at around 80% conversion).
Example 8: Table 14 shows typical reactor performance for the oxidative dehydrogenation of butane over a variety of monolithic catalysts at nitrogen dilution levels of 20% at total flow rate of 2 SLPM. Again, the less expensive chromium oxide catalyst shows similar selectivity and conversion trends with changes in the fuel to oxygen ratio as the Pt catalyst. In general, the chromium oxide catalysts have a higher conversion of the hydrocarbon and selectivity to C2H4 than the Pt catalyst. The chromium oxide catalysts produce less CO especially at the higher fuel to oxygen ratios.
Table 14: Carbon atom and hydrogen atom selectivity, conversion, and temperature for butane oxidation over various catalysts at several HK O? ratios with 20% dilution and a total flow rate of 2 SLPM iα an autothermal reactor at a pressure
Example 9: Table 15 shows typical reactor performance for the oxidative dehydrogenation of isobutane over a variety of monolithic catalysts at nitrogen dilution levels of 20% at total flow rate of 2 SLPM. Again, the less expensive chromium oxide catalyst shows similar selectivity and conversion trends with changes in the fuel to oxygen ratio as the R catalyst. In general, the chromium oxide catalysts have a higher conversion of the hydrocarbon and selectivity to C
2H
4 than the Pt catalyst. The chromium oxide catalysts produce less CO especially at the higher fuel to oxygen ratios.
Table 15: Carbon atom and hydrogen atom selectivity, conversion, and temperature for isobutane oxidation over various catalysts at several iC4Hlc/02 ratios with 20% dilution and a total flow rate of 2 SLPM in an autothermal reactor at a pressure of 1.2 at .
In addition to Pt-coated foam monoliths, Yokoyama, et. al. [Yokoyama, C. et al,
Catal Lett 1996. 38, 81-188; US Patent # 5905180] examined the addition of various metal promoters to Pt-coated foam monoliths for the autothermal oxidative dehydrogenation of ethane at millisecond contact times. Their study found that the
5 addition of Sn and Cu to Pt-monoliths enhanced the C2H selectivity and C^Hg conversion with no deactivation or volatilization of the catalyst. For the Pt-Sn catalyst, the ethane conversion increased by up to 6% and the ethylene selectivity increased by up to 5%. For the Pt-Cu catalyst, the conversion and ethylene selectivity were higher than Pt alone, but showed less improvement than the Pt-Sn catalyst.
I D Researchers have also examined oxidative dehydrogenation over oxide catalysts
[US Patent #4658074; US Patent #5430209; US Patent #5439859; US Patent #3862255; US Patent #5759946; US Patent #4026920]. These preocessβs prodice olefins with yields less than 45%.
The activity and selectivity of chromium oxide-supported catalysts in alkane
1 dehydrogenation has been known for many decades [Frβy, F. E. and Huppke, W. F ., Ind. Eng. Chem. 1933, 25, 54; Marciily, C. and Delmon, B., J. Catal. 1972, 24, 336-347]. Due to their importance in hydrogenation, dehydrogenation, and polymerization reactions, chromium-containing heterogeneous catalysts have been studied by a variety of techniques by a large number of researchers in the past several
2 decades. For the supported chromium Phillips polymerization catalyst, much of the research has focused on the structure and oxidation state of the reactive chromium surface species [Zecchina, A. et al, J. Phys. Chem. 1975, 79, 966-972; Weckhuyβen, B. M. et al, Chem. Rev. 1996, 96, 3327-3349]. It is now generally accepted that the catalytic properties of the supported chromium oxide are due to the surface chromium
25 species formed as a result of chromium-support interactions [Kim, O. S. and Wachs, I.
Example 10: The copper modified chromium oxide catalyst supported on a zirconia monolith shows excellent stability for the oxidative dehydrogenation of ethane even at high C2Hβ/02 ratios and high flow rates where other catalysts tend to deactivate. Typical reactor performance is shown in Table 16.
Table 16: Carbon atom and hydrogen atom selectivity, conversion, and temperature for ethane oxidation over a 45 ppi, 6.7% Cu - 6.7% CrjOj Zτθ2 monolith as a function of flow rate at several CJEtyOj ratios with 50% N2 dilution in an
Example 11: The oxidative dehydrogenation of ethane was investigated where the feed gas to the reactor contained CaH
β, 0
2, and H
2 with a small amount of N
2 for use as an internal calibration standard. The level of H
2 in the feed ranged from 0 to 50% that corresponded to H2 O2 ratios ranging from 0 to 3.0. The amount of N
2 in the feed was always less than 1 %. In the range of conditions studied, no flames were observed.
The addition of H
2 (Table 17) during the oxidative dehydrogenation of ethane over a 1.0% Cu - 9.0% Cr
∑O^lZrOz monolith catalyst with less than 1% N
2 dilution at a total feed flow rate of 7 SLPM results in a significant increase in the C
2H
4 selectivity. At a C
2H
β/O
2 ratio of 2.0, the selectivity to C
2H
4 increases from 56% with no H
2 addition to 76% with a Ha 0
2 ratio of 3.0 in the reactor feed. At a C
2Hβ 0
2 ratio of 2.4, the selectivity to C
2H
4 increases from 68% with no H
2 addition to 81% with a H2/O2 ratio of 3.0 in the reactor feed. The increase in the C2H4 selectivity corresponds with a decrease in the selectivity to CO and CO
2. The selectivity to CO and CO* falls proportionally with increasing H
a in the feed. The selectivity to CHU also falls slightly with increasing H
2 addition. The addition of H
2( however, also results in a decrease in the C
2H« conversion which falls from 94% to 81% at a C
2rVO
2 ratio of 2.0, and 60% to 66% at a
ratio of 2.4 as the H
2/O
2 feed ratio increases from 0 to 3. Along with the decrease in the C
2H
β conversion, the temperature measured at the exit of the catalyst at both C
2Hβ 0
2
ratios falls with increasing H
2 content in the feed. Thus, the overall production or yield of C
2H
4 in the reaction system is also affected by H
2 addition in the feed since the selectivity to C
2H
4 increases with H
2 addition and the C
2H
e conversion decreases with the addition of H
2. At a C
2Hβ 0
2 ratio of 2.0, the C
2H yield steadily increases with higher H2/O
2 ratios in the feed, but the marginal increase decreases at the higher H
2O
2 ratios. Thus, the selectivity to the desired product, C2H4, increases faster than the decrease in the C
2Hβ conversion with increasing t-fe concentration in the feed. At a C
2H
6/O
2 ratio of 2.4, the yield of C2H4 goes through a maximum at a H
2/O
2 of 1.0. This directly results from the fact that at a H
2/O2 below 1 , the C
2H selectivity rises faster than the corresponding decrease in C
2H
6 conversion. However, at a Hz/Oz ratio of 2 or above, the C2H4 selectivity rises slower than the decrease in the conversion of C
2H$.
Table 1 : Carbon atom and hydrogen atom selectivity, conversion, and temperature for ethane oxidation over various catalysis at several Cir Oj ratios with 20% N3 dilution and a total flow rate of 2 SLPM in an autothermal reactor at a pressure of 1.2 arm.
EXAMPLES
Example 1 ; The oxidative dehydrogenation of ethane was examined over a variety of catalysts supported on ceramic foam monoliths at a flowrate of 2 standard liters per minute (SLPM) which corresponds to a contact time of approximately 15 ms (GHSV = 200,000 hr
"1). The ceramic foam monolith supports (alpha-alumina or magnesia stabalized zirconia were obtained from Vesuvius Hi-Tech Ceramics and contain 45 ppi (pores per linear inch). Chromium was deposited on the support surface by exposing the ceramic to a liquid solution containing chromium ions such as aqueous chromium nitrate. Platinum was deposited on the support surface by exposing the ceramic to a liquid solution containing platinum ions such as aqueous chloroplatinic acid. Copper was deposited on the support surface by exposing the ceramic to a liquid solution containing copper ions such as aqueous copper nitrate. Between depositions, the catalysts were calcined in air.
Table 1 shows product carbon and hydrogen atom selectivity, conversion, and reaction temperature at several C
2H6 O
2 ratios in the feed gas with 20% N
2 dilution for 10 wt. % CraOa α-AlaOa, 9 wt. % Cr
2θ3 Zr0
2, 3 wt. % Pt/α-AfeOa, 6.7 wt. % Cu - 6.7 wt. % Cr/Zr0
2, and Pt (0.05 wt. %) modified 10 wt. % Cr
2θ3 α-AI
2θ3 catalysts. At a C
2Hβ/0
2 ratio of 1.2, the 9 wt. % Cr
203 AI
203 and Pt modified 0 wt. % Cr
2θ3/AI
2θ3 catalysts have a C
2H
4 selectivity around 20% compared to the 30% and 32% selectivity to C
2H4. seen for the 3 wt. % Pt/α-AfcOa and 9 wt. % Cr
20a/ZrO
2 catalysts respectively. The CO selectivity is approximately 40% for all the catalysts except for the Cr≥Oa ZrOa catalyst which showed a slightly lower selectivity to CO of approximately 35% and the Cu modified catalyst which showed a slightly higher selectivity to CO of -50%. For the selectivity to C0
2, a significant difference can be seen between the Cr
2θ3 and Pt catalysts with more CO
2 formed over the Cr
20
3 than over Pt. The C0
2 selectivity for the Pt-modified chromium oxide catalyst falls directly between the two single component catalysts. The CH4 selectivity, however, is essentially the same for ail the catalysts at approximately 13%. The selectivity to C-
2H
2 is slightly higher for the Cr
2θ
3 AI
2θ
3 and the Cu modified catalyst than the three other catalysts. At a C
2H
6 O
z ratio of 1.2, the highest selectivity to H
2 is about 56% for the Cr
2θ3 Al
2Oa and Pt-modified Cr
2OjyΑI
2θ3 catalysts. The H
2 selectivity for the CrgOai'ZrOa was 47% and the Pt Al
20
3 catalyst was 39%. The corresponding selectivity to H
2O is higher for the
catalysts than the two Cr
2θ3 Al
2θ
3 catalysts.
Similar trends between the catalysts exist at higher C2H602 ratios. The Cr2θ3 Zrθ2 catalyst and the Cu modified Cr2θ3/Zrθ2 catalyst show the beet C2H selectivity at the highest C2H6 O2 ratios with the Cu modified Cr2θ32rθ2 catalyst showing the lowest selectivity to CO. The C2H6 conversion and C2H yield is higher over the Cr2θ3 Zrθ2 catalyst and the Cu modified OtsOs/ZrOz catalyst than over the Pt
catalyst; the difference increases with increasing ratios of C2Hβ 02 in the feed. At C2H6/OZ ratios of 1.2 and 1.5, the C2H4 selectivity for the Cr2θ3 Zr02 and the PVAI2O3 catalysts is essentially the same with a lower C2H4 selectivity for the Cu modified Cr2θ3/Zrθ2 catalyst. But at the higher C2H6O2 ratios, the selectivity to C2H is higher over the chromium oxide and Cu modified chromium oxide catalysts at >70% compared to <65% over the Pt catalyst. The H2 selectivity for the Cr2θ3 Zr02 catalyst and the Cu modified Cr2Oy7r02 catalyst is also higher at all C2Hβ/O2 ratios, especially at the higher C2H6 O2 ratios. The CO selectivity is tower for the chromium oxide-zirconia catalyst and at the higher C2Hβ 02 ratios is approximately half of the CO selectivity for the Pt n catalyst. This reduction in CO production is even more evident at the higher C2H6/O2 ratios over the Cu modified Cr2θ3 r02 catalyst. However, the selectivity towards complete combustion products. C02 and HgO, is higher over the Cr2θ3/Zr02.
All catalysts exhibit similar C2H6 conversions with lower conversions at the higher C2H6/O2 ratios. At C2Hβ/02 ratios greater than 1.2, the chromium oxide containing
!5 catalysts all show a slightly higher C2He conversion than the Pt monolith. In ail cases, 02 is essentially completely consumed.
Example 2: The oxidative dehydrogenation of ethane was examined over a large variety of catalysts supported on ceramic foam monoliths at flow rates ranging from 2 to
:o 4 standard liters per minute (SLPM) which corresponds roughly to catalyst contact times of 5-15 ms (GHSV = 80.000 - 200,000 hr1).
Table 2 shows product carbon and hydrogen atom selectivity, conversion, and reaction temperature at a C2Hβ 02 ratio of 2.0 in the feed gas with 20% N2 dilution at a total feed flow rate of 2.0 SLPM for the oxidative dehydrogenation of ethane over single
25 component catalysts. The Cr2θ3, CuO, and MnOx catalysts on a Zr02 monolith have
Table 2: Carbon atom and hydrogen atom selectivity, conversion, and temperature for ethane oxidation over various an
essentially the same reaction results under these conditions with a C
2H selectivity of -68% with over 81% conversion of the ethane feed. These catalysts, Cr
2θ3, CuO, and MnO
x, also have very low selectivity to CO of 10%, 10%, and 7%, respectively; along with a corresponding higher C0
2 selectivity than the other single component catalysts. The 3 wt. % Pt α-AI
20
3 catalyst had a slightly lower selectivity to C
2H
4 and lower conversion of C
2Hβ than the Cr
20
3, CuO, and MnO* catalysts. Additionally, the Pt AI
2θ
3 catalyst also had a higher selectivity to CO and lower selectivity to CO
2 than the three transition metal oxide catalysts. The 9.1 wt. % FeO„Zrθ
2 catalyst also had a lower selectivity to C
2H (51%) and lower C
2H
β conversion (74%) along with a significantly higher selectivity to CO (31%) than the Crg
δ, CuO, and MnO
x catalysts. In contrast to all the other single component catalysts, the 9.7 wt. % Ni on a Zr0
2 monolith produced primarily syngas products, CO and H
2, and not oxidative dehydrogenation products,
Example 3: Table 3 shows the results for the oxidative dehydrogenation of ethane over Crj
j/ZrOs monolith catalyst with the addition of a single promoter including Ag, Cu, FβO
Xt Ni, MnO
x, and MgO. The reaction results for a C
2Hβ O
2 ratio of 2.0 with 20% N
2 dilution at a total flow rate of 2 SLPM show that the addition of the promoters to the Cr
2θ
3 Zrθ
2 catalyst can change the product selectivities and conversion over the catalyst. The addition of Cu or Ag to the Cr
2θ3 ZrO
2 catalyst does not significantly affect the product selectivities or reactant conversions. The addition of FeO* or MgO to the Cr
2O
3 ZrO
2 catalyst results in a slight decrease in the ethane conversion from 81% to 75% and 78%, respectively, but does not significantly affect the product selectivities. In contrast, the addition of MnOx to the Cr
2Oa Zrθ2 catalyst results in a decrease in the C
2H selectivity to 61 % from 68% and a corresponding increase in the CO selectivity to 22% from 13%. The ethane conversion or the 0.8 wt. % MnO
x - 8.0 wt. % Cr
20s/ZrO
2 catalyst, however, is the same as the 9.0 wt. % Cr
sθ3 ZrO
2.
Table 3: Carbon atom and hydrogen acorn selectivity, conversion, and temperature for ethane oxidation over various catalysts at a H«/0_ rauo of 2.0 with 20% N3 dilution and a total flow rate of 2 S PM in an autothermal reactor at a im.
Example 4: Table 4 shows the results for multi-component catalysts for the oxidative dehydrogenation of ethane for a C2Hβ 02 ratio of 2.0 with 20% dilution at a total flow rate of 2 SLPM. The addition of MgO, FeOx, and MnOx to a Cu/Cr2θ3/Zrθ2 results in a lower selectivity to C2H4 and corresponding higher selectivity to CO along with a lower CaHβ conversion than the 1.3 wt. % Cu - 13.0 wt. % Cr2θ3 on a Zr02 monolith. The addition of Ni to the Cu/Cr2θ3/Zr02 catalyst also resulted in a slightly lower C2H4. selectivity, but the addition of Ni did not significantly affect the CaHβ conversion.
The addition of the Cu to the Cr203/Zr02 catalyst does not significantly affect the product selectivity or rβactant conversions as noted above. The addition of Cu. however, did significantly increase the activity of the CuO CraOs Zr k catalyst especially at higher flow rates, N2 dilutions, and C2I θ2 ratios. The 1.3 wt. % Cu - 13.0 wt. %
Table 4: Carbon atom and hydrogen atom selectivity, conversion, and temperature for ethane oxidation over various catalysts at a CΑJOi ratio of 2.0 with 20% N_ dilution and a total flow rate of 2 SLPM in an autothermal reactor at a ressure of 1.2 atm.
Cr
20
3 Zr0
2 catalyst was able to achieve stable operation at a C
2Hβ 0
2 ratio of 2.4 with 50% dilution at a total flow rate of 10 SLPM with no signs of deactivation. In contrast, the 9 wt. % Cr
2θ
3 Zr0
2 catalyst was only able to achieve stable, non-deactivating behavior at a C
2Hβ 02 ratio of 2,4 with 50% N
2 dilution at a total flow rate of less than 4 SLPM. It was also found that the addition of the Cu promoter to a MnO
x Zr0
2 catalyst did not result in the same dramatic increase in catalyst activity.
Example 5: Under some conditions, reaction is not sustained indefinitely over some of the supported chromium oxide catalysts, The onset of extinction depends on the composition of the catalyst, the CzH^lOz ratio, the support material, the weight loading of the catalyst, the flowrate, the N2 dilution, and the preheat of the feed.
The time to extinction is shown in Table 5 as a function of C2Hβ/02 ratio with 20% reactant dilution at a flowrate of 2 SLPM for the various catalysts. The sustained activity of the Cr2θ3/AI2θ3 and the Pt Al203 and Cr2θ3 Zr02 catalysts differ in that the Pt Alz03> Cu modified Cr2θ3 Zr02. and the 9 wt. % Cr203/Zr02 catalyst sustain reaction for a much longer period of time than the other catalyst compositions examined. Like the Pt monolith, the 10 wt. % Cr2θ3/AI2θ3 catalyst at a C2rV02 ratio of 1.2 with 20% N2 at a flowrate of 2 SLPM has shown sustained reaction for very iong periods of time with no deactivation and product selectivity changes. However, at a C2rV02 ratio of 1.5, the reaction is sustained for only 1.6 hours. Before extinction of the reaction, the product selectivity and reactant conversions are constant over the entire time period. The catalyst temperature falls slightly in the first 1.25 hours and then rapidly decreases until the reaction extinguishes. At a C2H6 02 ratio of 1.8 with 20% N≥ dilution at 2 SLPM, the reaction over the Cr2θ3 Al2θ3 monolith extinguishes after only 5-10 minutes.
Table 5; The reactive lifetime of the various catalysts as a function of C2r O. ratio in the feed with 20% N. dilution with a total feed flow rate of 2 SLPM in an autothermal reactor at a pressure of 1.2 atm.
Fuel/02 1.2 1.5 1.8 2.0 2.2 2.4
Catalyst Lifetime <hrs)
3 % Pt/α -AI2O3 >5,0* >5.0* >5.0* >S.O* >5.0* >5.0*
9% CriOj/ZrO; >2.0* >2.0* >6,0* >2.0* >2.0* >2.0*
2* Cr203Zrθ2 >2.0* >2.0* 0.1 — — —
10% Crj α-Al >4.0* 1.6 0.1 — — —
1.5% CrjCVα -AI2O3 0.1 <0,1 — — — —
6.7% Cu .6.7% C^CyZrO, > 2.0* > 1.0* >10* > 1.0* > 1.0* >2.0*
10% CrjOj/α -Al.O. Pellets > 1.0* >7.5* 0.2 0.2 — .... j 1.3 % Cu - 13.0% Crιθ3/Zr02 > 1.0* >ι.o* >1.0* > 1.0* > 1.0* >5.0*
9.1%FeO Zrθ3 >1.0* >1.0* >1.0* > 1.0* > 1.0* >1,0*
8.1 MnO-/Zτθ2 > 1.0* >1.0* > 1.0* > 1.0* > 1.0* >1.0*
9.7 % NiZrθ2 > 1.0* >1.0* >1.0* > 1.0* > 1.0* >1.0*
11.2%Cu0/2rO2 >1.0* >1.0* >1.0* > 1.0* > 1.0* >1.0*
PM0% CrjO α -AlαO? >1.0* >7.0* 9.0 ... — ...
* Cataiyst did not show any signs of deacαvation
In contrast to the Cr203/AI2O3 catalyst, the reaction was sustained under these conditions for the 9 wt. % Cr2θ3/ r02 catalyst and the Cu modified Cr2θ3Zrθ2 catalyst for longer periods, similar to the Pt/Al203 monolith catalyst. Like the Pt catalyst, a 9 wt. % Cr2θ3Zr02 catalyst showed non-deactivating steady state behavior up to a C2H6O2 ratio of 2.4 with 20% N2 dilution at a total flowrate of 2 SLPM which is well outside of the flammability range for this system. At a C2Heθ ratio of 1.8 with 20% N2 dilution at 2 SLPM, the 9 wt. % Cr2θ3ZrO2 monolith catalyst did not show any deactivation after 5 hours of continuous operation, while the reaction quickly extinguished under these condition for lower chromium oxide loadings on Z1O2 and on all Craθ3AI203 catalysts. At a C2Hβ/02 ratio of 1.5 with 50% N2 dilution at 2 SLPM, the Cu modified Cr203/Zr02 monolith catalyst did not show any deactivation for over 7 hours of continuous operation. The chromium oxide loading of the monoliths can significantly affect the catalysts. For chromium oxide supported on Zr02 and Al203, the activity of the catalyst increases with the chromium oxide loading.
The other transition metal oxide catalysts (11.2 wt. % Cu/Zr02, 9.7 wt. %
Ni/Zr02, 9.1 wt. % FeO*/Zr02, and 8.1 wt. % Mn0χ/Zr 2) are able to sustain reaction for more than one hour under these conditions. Additionally, the addition of the promoters (Ag, Ni, Mn, Mg, Fe, Cu, their oxides, and combinations thereof) to the Cr2θ Zrθ2 led to stable operation for several hours with no signs of deactivation at a C2H6/O2 ratio of 2.4 with 20% N2 dilution at a total flow rate of 2 SLPM.
The addition of very small amounts of Pt (> 0.001 g , -0.05 wt. %), as a co- catalyst, was examined for its influence on the ability of the supported chromium oxide catalysts to sustain reaction for extended periods. Table 5 shows the time to extinction for Pt-modified C^Os/AI≤Oa and unmodified Cr20*'AI203 monol'rth catalysts at several C2H6 02 ratios with 20% N2 dilution at a flowrate of 2 SLPM. The activity of Pt-coated monoliths for the oxidative dehydrogenation is relatively independent of the Pt loading, even for very small amounts of Pt [Haubein, N., Effects of Support Material and Pt Loading on the Catalytic Oxidative Dehydrogenation of Ethane; University of Delaware, 1997]. The addition of Pt significantly increases the length of time that reaction can be sustained over the Cr2θ3/AI2θ3 catalyst. However, the Pt-modified Cr2θ3 Al203 catalyst still extinguished rapidly at the higher C2Hβ/02 ratios unlike the Pt Al203 catalyst.
The surface area can also influence the temporal behavior of the catalyst. The surface area of the monolith supports is approximately 200 cm2 per gram, which is approximately 400 cm2 of total surface area for the monoliths used here. The total surface area can be increased by replacing the monolith with a packed bed of 3 mm Al203 pellets. The surface area of blank a-AfeOa pellets is 9 m2 per gram which results in a catalyst bed with approximately 27 m2 of total surface area. The packed bed contained 60 Cr2θ3 AI203 pellets, which was approximately equal in total weight to the chromium oxide monolith catalyst and resulted in an approximate bed depth of 1.5 cm.
Table 6: The reactive lifetime of the various catalysi s as a function of the total feed flow rate at several C2H«/0^ ratios m the feed with 20% N. dilution in an autothermal reactor at a pressure of 1.2 atm.
F eL/Oj 1.2 1.2 1.2 1.3 1.5 1.5 1.8 1.8 1.8
Flowrate (SLPM) 1.0 2.0 3.0 1.0 2.0 3.0 1.0 2.0 3.0
Catalyst Lifetime (hrs)
3 % Pt/α-Al3θ3 > 2.0* > 5.0* > 2.0* > 2.0* > 5.0* > 2.0* > 2.0* > 5.0* > 2.0*
10% α2Oj/ α-Al2θ3 > 1.0* > 4.0* < 02 7.0 1.6 < 0.1 7 0.1 — l ,5% Cr20ιt α -Al20} > 6.5 0.1 — 0.1 < 0.1 — — —
6.7% Cu - * * * * * *
6.7% Cr203 Zr02
9% QI J/ZΓQJ > 2.0* > 2.0* > 6.0* > 3.0* > 2.0* > 3.0* > 1.0* > 6.0* 2.8
4% > 2.0* > 2.0* > 5.0* > 2.0* > 9.0* 1.5 7
2% CriOj/ZrOi * > 2.0* 7 ? ? ? 0.1 —
Catalyst did not show any signs of deacuvation
Reaction was sustained over the 10 wt. % C^Oa/α-A Os pellets for more than about 7.5 hours with no sign of deactivation or changes in the product selectivity at a C2H6 02 ratio of 1.5 with 20% N2 dilution at a flowrate of 2 SLPM. Under the same conditions, the 10 wt. % Cr2θsAI203 monolith extinguished in 1.6 hours. In addition to the difference in reactive lifetimes, the chromium oxide pellet catalyst displayed a slightly higher selectivity to C2H4 and lower selectivity to CO. The chromium oxide pellets also displayed a slightly lower C2H6 conversion, 94%, compared to 96% CsHβ conversion over the chromium oxide monolith. Therefore, in addition to the composition of the ceramic support, the total surface area of the catalyst can also have a significant impact upon the behavior of the supported chromium oxide catalyst.
The feed flowrate strongly affects the sustainability of reaction over the chromium oxide catalysts as shown in Table 6. For the 10 wt. % Cr2θ3/Al2θ3 catalyst at a C2Hβ 02 ratio of 1.5 with 20% N2 dilution, the reaction was sustained for 7 hours at 1 SLPM, 1.6 hours at 2 SLPM, and only a couple of minutes at a flowrate of 3 SLPM. When the chromium oxide loading is reduced, the activity of the catalyst decreases accordingly. For a 1.5 wt. % Cr2Os/α:-Al2θ3 monolith, reaction at 2 SLPM with 20% N2 dilution could not be sustained even at a C2Hβ/θ2 ratio of 1.2. The total flowrate had to
be reduced by 50% to 1 SLPM before reaction over the low chromium oxide containing catalyst was sustained for an extended period of time (>6 hrs.) In spite of the lower flowrate and hence longer contact time, the catalyst did not achieve steady state. Throughout the time survey, the catalyst showed signs of deactivation with the conversion of C2H6 and 02 falling with time on stream.
For the Cr2θ3 Zrθ2 catalysts, the flowrate also strongly affects the catalyst's lifetime. At a C2H6 O2 ratio of 1.5 with 20% N2 dilution over the 4 wt. % Cr2θ9 Zr02 monolith catalyst, the reaction was only sustained for 1.5 hours at a flowrate of 3 SLPM, compared to over 9 hours at 2 SLPM. Similar decreases in the lifetime with increasing flowrate, and correspondingly shorter contact time, can be seen for the 9 wt. % Cr2θ3 Zrθ2 catalyst at the higher C2H6/O2 ratios. The Cu modified Cr^/ZtOz catalyst showed no signs of deactivation for C2Hβ/02 ratios up to 2.4 and total flow rates up to 4 SLPM.
In addition to the lifetime of the catalyst, the flowrate affects the product selectivity and ethane conversion shown in Tables 7 and 8. As the flowrate increases, the selectivity to C2H4 decreases and the selectivity to C2H2 increases; the H2O selectivity decreases; and the conversion increases slightly. The most significant effect of increasing the flowrate can be observed by examining the temperatures measured at the front and back of the catalyst. At the lowest flow rates, the temperature on the front face of the catalyst exceeds the temperature at the rear face of the catalyst; while at the higher flow rates, the front temperature is much less than the rear temperature.
The lifetime of the chromium oxide catalysts is also influenced by the N
2 dilution in the feed mixture. At 50% N
2 dilution at 2 SLPM with a C
2Hβ/0
2 ratio of 1.2, the reaction over the 10 wt. %
catalyst was only sustained for 20 minutes before the reaction rapidly extinguished, and at a C
2Hβ θ
2 ratio of 1.5, the reaction