This invention relates to the cracking of higher hydrocarbons into alkenes using the hydrogen form of pentasil type zeolite catalysts (HZSM-5). In particular, the invention relates to the use of HZSM-5 catalysts having specific silicon to aluminium ratios in the cracking of hydrocarbons to give high yields of propylene.
Zeolites are a well-known class of aluminosilicates containing an (Si, Al)nO2n framework with negative charge balanced by cations present in the framework cavities. The nature of the cations used can vary greatly but typical examples include potassium, sodium, calcium, silver and hydrogen. The spaces created within a zeolite framework are relatively large and these so called “cages” can accommodate large molecules. However, the external pores of the framework which allow access to the cages of the zeolite tend to be smaller, allowing control over the sizes of molecules which can enter and leave the zeolite structure.
The possibility of size selectivity has made use of zeolite catalysts an attractive option for the skilled chemist when carrying out cracking reactions, i.e. reactions where higher hydrocarbons are broken down into smaller hydrocarbon fragments, as well as other reactions such as isomerisation reactions or aromatisation reactions.
The cracking of higher hydrocarbons into smaller hydrocarbons, in particular ethylene and propylene, is well known and has been carried out in the petrochemical industry for many years. For example, the use of a steam cracker to convert higher hydrocarbons by pyrolysis into ethylene and propylene is well known. In general however this reaction produces much more ethylene than propylene. Zeolite catalysis therefore provides a further cracking alternative for the skilled chemist.
U.S. Pat. No. 6,222,087 describes the use of phosphorus doped ZSM-5 and/or ZSM-1 zeolites in the formation of light olefins rich in propylene from a hydrocarbon feed containing C4-7 olefins/paraffins. With a 1-butene feed the invention realises 37% ethylene and propylene but with a light catalytic naphtha feed only 25% ethylene and propylene is formed.
U.S. Pat. No. 5,968,342 describes the use of various metal doped zeolites as catalysts in the conversion of naphtha to ethylene and propylene. ZSM-5 catalysts in which the proton is exchanged with silver or copper are suggested to ensure that the zeolite is substantially free of protons. The text suggests that HZSM catalysts show poor selectivity for lower olefins. GB 2345294 describes a process similar to that in U.S. Pat. No. 5,968,342, i.e. one in which a proton free zeolite is employed in a cracking reaction.
WO 00/18853 describes ZSM-5 catalysts doped with phosphorus and a further promoter such as tin or gallium for use in cracking naphtha to olefins.
With the increasing demand for propylene throughout the world, in particular the increasing demand for polypropylene, it would be useful if more of the heavier petrochemical fractions could be converted efficiently into propylene for polymerisation.
Such heavier petrochemical fractions are often rich in aromatic compounds notably benzene. The use of aromatic compounds in fuels is being reduced for environmental reasons resulting in a potential surplus of aromatic compounds. Moreover, whilst there is a market for benzene, the market is relatively small. Thus, it is anticipated that the market for aromatic compounds may become saturated. It would therefore be useful if aromatic compounds such as benzene could themselves be converted in high yield into more marketable olefins such as ethylene and propylene.
There remains therefore, the need for further cracking processes to be developed to maximise the yields of ethylene and propylene which can be obtained from hydrocarbon feeds. In particular it would be useful if the amounts of propylene produced could be maximised. It has now been surprisingly found that by using a hydrogen ZSM-5 catalyst with a particular silicon to aluminium molar ratio hydrocarbons, such as hydrogenated aromatic fractions or hydrocarbon fractions containing higher olefins, can be cracked into ethylene and propylene in very high yields. Contrary to the teachings of the prior art, these catalysts show excellent olefin selectivity at the Si/Al molar ratios claimed and can be used to crack a variety of feeds to give rise to olefins such as ethylene and propylene in high yield.
Thus, viewed from one aspect the invention provides a process for cracking C4+ hydrocarbons comprising heating the hydrocarbons at a temperature of 400-800° C. in the presence of an H-ZSM-5 catalyst having a silicon to aluminium ratio of 61-1000.
Viewed from another aspect the invention provides the use of an HZSM-5 catalyst having a silicon to aluminium ratio of 61-1000 as a catalyst in the cracking of C4+ hydrocarbons.
The process of the invention may be used on its own or in combination with other processes such as alternative olefin producing processes, e.g. pyrolytic steam cracking.
By C4+ hydrocarbons is meant that the hydrocarbons being cracked have at least four carbon atoms. Preferably, the hydrocarbons being cracked will have between 4 and 20 carbon atoms, preferably between 5 and 12 carbon atoms, e.g. 6 to 10 carbon atoms. The hydrocarbon feedstock can be a single pure hydrocarbon but more usually it will be a mixture of various hydrocarbons, e.g. a light or heavy naphtha fraction, different condensate fractions or a hydrogenated pyrolysis gasoline. The hydrocarbons may be saturated or unsaturated, linear, branched or cyclic and preferably non-aromatic. Preferred hydrocarbons however will be saturated paraffins (alkanes such as pentane, hexane, octane, decane, dodecane), olefins of general formula CnH2n where n is from 4 to 12 or most preferably saturated cyclic hydrocarbons, e.g. cyclopentane, cyclohexane, methylcyclohexane, dimethylcyclohexane, methylcyclopentane or decalin. The hydrocarbon feedstock may also be a mixture of any of the above. In a highly preferred embodiment the feedstock comprises at least one C4+ olefin, e.g. a C5+ olefin or diene.
The hydrocarbon feed may contain aromatic compounds such as benzene. Such aromatics do not, in general, crack in the presence of the HZSM-5 catalyst of the invention and are therefore left unchanged. Aromatic compounds may therefore be used as diluents in the process of the invention. Preferably however, it would be useful to convert aromatic compounds into useful olefins.
Many of the potential feeds to the catalytic cracker contain high volumes of aromatic compounds, e.g. pyrolysis gasoline or condensate fractions direct from an oil refinery or an oil producer.
In a preferred embodiment therefore, the hydrocarbon feed may be hydrogenated prior to exposure to the zeolite catalyst. Thus for example, a hydrocarbon feedstock containing benzene can be hydrogenated to give cyclohexane prior to cracking. By hydrogenating the feedstock, the majority of the hydrocarbons fed to the cracking reactor will be saturated, i.e. alkanes and cyclic saturated hydrocarbons. This forms a further aspect of the invention.
Thus viewed from a further aspect the invention provides a process comprising hydrogenating a C4+ hydrocarbon feedstock; and heating the resulting hydrocarbons at a temperature of 400-800° C. in the presence of an H-ZSM-5 catalyst having a silicon to aluminium ratio of 61-1000.
A potential reactor set up could involve diluent free cracking, e.g. using a hydrogenated pyrolysis gas feed with recycling of C4+ fractions to the hydrogenator. Alternatively, a reactor set up could involve the use of an aromatic compound as a diluent for a hydrocarbon feedstock. Pyrolysis gasoline from a naphtha steam cracker, which contains a large amount of aromatic compounds, would, for example, be suitable as such a diluent.
The cracker may then have a recycling system which passes through a hydrogenator to convert the diluent into a crackable hydrocarbon. This may then be fed back into the cracker (along with fresh diluent) so that in a first pass, the aromatic compound acts as a diluent prior to hydrogenation and itself being cracked. Recycling of C4+ fractions could occur simultaneously.
Preferably the temperature within the cracking reactor should range from 500-750° C., more preferably 550-700° C., especially 550 to 650° C., e.g. 570° to 650° C. or 570 to 630° C. such as about 600° C. The residence time of the feed over the catalyst should be long enough to give substantial conversion of the feed but not long enough to give rise to the production of large percentages of aromatic compounds.
It has been surprisingly found that when the process described above is employed the yield (calculated as weight of carbon) of ethylene and propylene can exceed 40 wt % carbon. Thus, 40% by weight of the carbon in the hydrocarbon feedstock is recovered as ethylene and propylene. Preferably at least 50 wt % carbon is recovered as ethylene and propylene.
Whilst the ratio of produced propylene to ethylene can vary, e.g. more ethylene than propylene or a 1:1 ratio, preferably, the bulk of the produced product is propylene e.g. at least 25 wt % C, especially 30 wt % C. It is also preferred if the ratio of produced propylene to ethylene is greater than 1.2:1, e.g. greater than 1.5:1.
An important factor in ensuring high percentage yields of ethylene and propylene is ensuring high conversion of initial feedstock. It is preferred if at least 50% of the initial feedstock is cracked, especially at least 60%, e.g. 70%, most especially at least 80% or at least 90%. At higher temperatures, conversion can near 100%. The higher the cracking temperature the higher the conversion. Also at higher temperatures, ethylene selectivity increases. There is, however, a trade off between temperature and hence high conversion/desired selectivity and potential for formation of aromatic compounds. The most preferred temperature to maximize conversion but minimise benzene formation is in the range 570 to 630° C.
Viewed from another aspect the invention provides a process for cracking C4+ hydrocarbons comprising heating the hydrocarbons at a temperature of 400-800° C. in the presence of an H-ZSM-5 catalyst having a silicon to aluminium ratio of 61-1000 and recovering the cracked hydrocarbons;
wherein at least 40% wt carbon of the cracked C4+ hydrocarbons is recovered as ethylene and propylene.
Any non-converted feedstock can of course be recycled back into the cracker along with any unwanted side products. The cracking reaction obviously yields hydrocarbons other than ethylene and propylene. Other products may include ethane, propane, methane, butane, butene and amounts of C5 and C6+ fractions.
Catalytic cracking reactors of use in the invention are known and can operate under the temperatures discussed above using pressure if necessary, e.g. from 0.1 to 10 atm, preferably 0.3 to 2 atm. The catalytic process can be carried out in, for example, a fixed bed, moving bed, or fluidised bed reactor and the hydrocarbon flow can be cocurrent or countercurrent to catalyst flow.
The catalyst may be formed from fine solid particles having a size range of from about 0.01 to 10 mm, e.g. 0.2 to 5 mm. Diluent such as an inert gas (nitrogen), methane or aromatic compounds can be employed as is known in the art, e.g. to carry the hydrocarbon gas stream into the reactor. The ratio of inert gas (if used) to hydrocarbons may range from 0:1 to 1000:1. Careful selection of diluents may allow further control over the ratio of products formed. Comprehensive discussions of suitable reactor set up can be found in the prior art cited above and are known in the art.
The catalyst employed in the invention, i.e. a HZSM-5 catalyst having a particular Si/Al molar ratio is obtainable from commercial sources such as Sud-Chemie. Silicon to aluminium ratios given in the text are molar ratios, i.e. by a Si/Al ratio of 100 is meant that the molar amount of Si is 100 times the molar amount of Al. The manufacture of ZSM-5 catalysts is described in U.S. Pat. No. 3,702,886. Preferred silicon to aluminium ratios are 80-400, especially 100-300, e.g. 85 to 200. It has been surprisingly found that at these high ratios, the amounts of formed lower olefins are high, in contrast to products obtained when using HZSM-5 catalysts having lower Si/Al ratios. The catalyst is preferably medium pore (e.g. 10 rings).
Comprehensive discussions of the manufacture and use of ZSM-5 catalysts can be found in the prior art cited above. The amount of catalyst employed will vary depending on the size of the reactor and the size of the feed but will be readily determined by the artisan. Catalyst can be continuously added to the cracker if necessary. It may also be necessary to regenerate the catalyst using known conditions. It has however, been surprisingly found that the catalyst of the invention only exhibits slow loss of activity during the cracking reaction and hence regeneration may be required only infrequently.
It is common for the zeolites to be calcined prior to use, however it is preferred if the HZSM-5 catalyst used in the present invention is not calcined. Moreover, many zeolite catalysts are aged prior to use by exposing them to hydrothermal treatment, e.g. heating at a temperature of 500 to 800° C. in the presence of steam. The zeolites used in the present invention should not be heated in this fashion prior to use since this may affect their light olefin selectivity, i.e. it is not necessary to expose the zeolite catalysts of the invention to elevated temperatures and 100% steam before use.
The catalyst of use in the invention is an HZSM-5 species as hereinbefore described. Hence, the cations within the zeolite should be protons. There is the possibility however that the zeolite may be contaminated with other cations, e.g. sodium or potassium ions. It is preferred if the zeolite of use in this invention comprises primarily protons with low amounts of other cations, e.g. exclusively protons and no other cations. For example, the level of cations (other than protons) should be less than 0.05% wt, preferably less than 0.01% wt.
It may be necessary to control the contact time of the hydrocarbon materials with the catalyst by taking pyrolysis properties of the hydrocarbons and reaction temperature into consideration. Contact times, measured as 1/GHSV (gas hourly space velocity) should be from 0.00002 h to 0.002 h.
The products of the cracking reaction can be fractionated using known techniques. Any unwanted products may be recycled to the reactor so as to increase yield of the desired product(s). Larger side products may be channeled into a condenser prior to recycling.
A suitable reactor set up is now described. The setup used in the first Example is shown schematically in
A larger scale apparatus is shown in
In
The top stream from separator (15) is itself passed to a further separator (19) via pump (18) where C1-3 fractions are separated from C4/5 fractions. These latter fractions are recycled to the cracker and the light fractions isolated as the desired product.
The invention will now be described with reference to the following non-limiting examples and Figures.
The reactor set up depicted in
The gas chromatograph used was an Agilente micro GC with four columns. Total analyses of permanent gases and C1-C6 took 240 seconds. The gas chromatograph was calibrated with a C1-C4 gas mixture (Standard 1) and a mixture of N2, H2 and C1-6 (Standard 2). Cyclohexane (CH) was calibrated by pass analysis. All C5 compounds were assumed to have the same calibration factor and all C6 compounds were assumed to have the same calibration factor.
Reactions were conducted in the temperature region 400-600° C. with 1 g of catalyst, 50 ml/min N2 flow through evaporator and 0.1 ml/min hydrocarbon flow corresponding to a WHSV [weight hourly space velocity] of 4.7 gCH/(g catalyst*h) with cyclohexane used as feed. The catalysts were pressed to tablets and then crushed into particles with particle size 0.2 to 0.5 mm before testing.
The reactor was first heated to 400° C. under nitrogen flow before switching to feed from the evaporator. Samples were taken from the reactor at 4-15 minute intervals. N2 was then flushed through the reactor and the reactor heated to 450° C. before feed from the evaporator was again admitted and the process above repeated. This process was repeated at 500° C., 550° C. and 600° C. before the reactor was cooled to 400° C. or 450° C. for gas chromatographic runs to be performed to check for any catalyst deactivation which may have occurred over the course of the experiment.
Four catalysts were examined using the above protocol:
Catalyst A: Kristal 232 ST (Grace Davison)—a standard cracking catalyst based on rare earth exchanged Y zeolite which has wide pores (12 rings)
Catalyst B: Valfoor CP 811 BL-25 (PQ) a Beta zeolite with wide pores (12 rings)
Catalyst C: A HZSM-5 with Si:Al=28 (Sud Chemie) Medium pores (10 ring)
Catalyst D: A HZSM-5 with Si:Al=85 (ZPO 31170) (Sud Chemie) (10 ring)
Catalysts A to C are comparative, catalyst D exemplifies the invention.
Cracking Catalyst A
The results for this catalyst are shown in
Catalyst B
The results for catalyst B are presented in
Catalyst C
The results are presented in
Catalyst D
The results for catalyst D are presented in
General Conditions
*Hydrogenated pyrolysis gasoline light fraction <125° C.
** Hydrogenated pyrolysis gasoline.
HZSM-5 from Sutd Chemie AG (ZPO31400) with a Si/Al=200 was pressed into tablets and then crushed into particles with particle size 0.2-0.5 mm. One gram of catalyst (1 g) particles was tested at 400-650° C. in a quartz fixed bed reactor with on-line GC analysis. A cyclohexane:N2 molar ratio of 1:2.2 and a GHSV of 4338 per h was used. Liquid cyclohexane (CH) from Merck (99.6%) was evaporated at 75° C. into the N2 stream before entering the reactor.
The conversion is calculated on basis of unconverted feed (nitrogen is used as internal standard) and the carbon selectivity* in the effluent to propene and ethylene is shown in Table 1.
*Carbon selectivity calculated based on the analysis of C1-C6 products
The catalyst as described in Example 2 was tested with cyclohexane using the same experimental procedure as described in Example 2. The catalyst was tested for 7105 minutes at 600° C. The conversion as a function of time on stream is shown in
The carbon selectivity to ethylene and propene (based on analysis of products C1-C6) is given in
The catalyst tested in Example 3 was regenerated in 5% oxygen in He. After regeneration the catalyst was tested with CH using the same experimental procedure as described in Example 2. The catalyst was tested for 1397 minutes at 600° C. The conversion as a function of time is shown in
The catalyst as described in Example 2 was tested with cyclopentane (CP) using the same experimental procedure as described in Example 2. A CP:N2 molar ratio of 1:1.9 and a GHSV of 4542 h−1 was used. Liquid CP from Janssen Chimica 16.775.91 (98%) was evaporated at 45° C. into the N2 stream before entering the reactor.
The conversion calculated on basis of unconverted feed (nitrogen is used as internal standard) and the carbon selectivity* in the effluent to propene and ethylene is shown in Table 2.
*Carbon selectivity calculated based on the analysis of C1-C6 products
The catalyst as described in Example 2 was tested with MCP using the same experimental procedure as described in Example 2. A MCP:N2 molar ratio of 1:2.7 and a GHSV of 4131 h−1 was used. Liquid MCH from Venton 12548 (99%) was evaporated at 90° C. into the N2 stream before entering the reactor.
The conversion calculated on basis of unconverted feed (nitrogen is used as internal standard) and the carbon selectivity* in the effluent to propene and ethylene is shown in Table 3.
*Carbon selectivity calculated based on the analysis of C1-C6 products
Hydrogentated pyrolysis gasoline was obtained from Statoil. Some of the hydrogenated pyrolysis gas was distilled to boiling point 125° C. This corresponds to 69.8 wt % of the feed. 98.7 wt % mass balance was obtained during the distillation (this may indicate that 1.3 wt % of the lightest components have been lost during the distillation). The total hydrogenated pyrolysis gasoline (Example 8) and the light fraction <125° C. (Example 7) were used as feeds.
Piona analyses of the two feeds were performed and the results are given in Tables 5 and 7 below. The total feed contains 80.1 wt % naphthenes and the light feed contains 83.2 wt % naphthenes. The density of the two feeds was measured by weighing 25 ml of the samples. The light fraction <125° C. had a density of 0.744 g/ml and the total hydrogenated pyrolysis gasoline had a density of 0.776 g/ml.
The analysis of the effluent was performed with an Agilente micro GC with 4 columns. All detectors were TC detectors
Column A: 5 A mol sieve
Run at 30.7 psi and 85° C. with Ar as a carrier gas.
Used for analysis of hydrogen and nitrogen
Column B: Poraplot U
Run at 20.7 psi and 70° C. with He carrier gas.
Used for analysis of methane, ethane and ethylene
Column C: Alumina Plot
Run at 34.3 psi and 115° C. with He carrier gas
Used for analysis of propene, propane and C4's
Column D: OV-1
In order to analysis more of the heavy compounds on the GC, the temperature of the D column was increased to 150° C., compared to 70° C. in earlier analysis. Pressure was 29.6 psi. He carrier gas. This column was used to analyse C5-C9 compounds.
The catalyst as described in Example 2 was tested with the light fraction boiling <125° C. of a hydrogenated pyrolysis gasoline using the experimental conditions as described in Example 1. A Piona analysis of this feed is given in Table 5 below. The yields to ethene and propene are given Table 6.
The sum of the C2= and C3=yields are about 43% at 650° C. After running the experiment for approximately 8 h, the catalyst has been exposed to feed for approximately 115 min. The conversion at 450° C. was about the same at the end of the experiment as in the start of the experiment, which points to no significant deactivation.
The catalyst as described in Example 2 was tested with a hydrogenated pyrolysis gasoline using the experimental conditions as described in Example 1. A Piona analysis of this feed is given in Table 7 below. The yield to ethene and propene are given Table 8.
The catalyst described in Example 2 was tested at 400-650° C. in a quartz fixed bed reactor with online GC analysis. Nitrogen and benzene were used as diluent, along with a cyclohexane:N2:benzene molar ratio of 1:1.3:1, and at a GHSV 4356 per h. Liquid 1:1 molar mixture of cyclohexane (CH) from Merck (99.6%) and benzene from Merck (p.a.) was evaporated at 80° C. into the nitrogen stream before entering the reactor.
In
*0.1 ml/min CH (liq) + 50 ml/min N2,
**0.1 ml/min CH (liq) + 0.098 ml/min Bz (liq) + 27.7 ml/min N2.
HZSM-5 from Sud Chemie AG (ZPO31400) with a Si/Al=200 was pressed into tablets and then crushed into particles with particle size 0.2-0.5 mm. One gram of catalyst (1 g) particles was tested at 500-700° C. in a quartz fixed bed reactor with on-line GC analysis.
C4 mix obtained from a commercial steam cracker 19 ml/min and 25 ml/min nitrogen was used as feed. The C4 mix was contained as a liquid in a container keeping a pressure of 1.5 bar. This was taken from a stream where the original C4 mix had been exposed to a partial hydrogenation of butadiene, and thereafter the isobutene removed by reacting to methyltertbutylether. The analysis of the C4 mix was performed on a separate GC and the analysis is given in Table 10.
The conversion of the C4's and the yield to C2=/C3= and aromates are given in
The catalyst (1 g) as described in Example 2 was tested with the light fraction of pyrolysis gasoline as feed. The pyrolysis gasoline was distilled and the fraction boiling at ≦125° C. was used as feed. This corresponds to 82% of the pyrolysis gasoline sample. Analysis of the light fraction of the pyrolysis gasoline was performed on a separate GC and the analysis is given in Table 11. This light fraction had a density of 0.798 g/ml.
The liquid feed was fed with 0.1 ml/min and evaporated into a 50 ml/min nitrogen stream before entering the reactor.
The carbon distribution in the gas effluent is given in
Number | Date | Country | Kind |
---|---|---|---|
0414442.4 | Jun 2004 | GB | national |
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/EP05/06927 | 1/5/2006 | WO | 5/5/2007 |