Trace Acetylene Removal from Ethylene
Robin L. McNeely
Louisiana State University
Ethylene is one of the most widely produced petrochemicals in the world, with a production history reaching back to the early 20th century. It has been recovered since 1930 as a product of coke-oven gas, and in the 1940s became an important industrial intermediate when American companies began producing it by separation from waste gas. This era also saw deliberate ethylene production from ethane and natural gas. More recently, ethylene has taken the place of acetylene in virtually all large-scale chemical syntheses. However, acetylene itself is a byproduct of modern ethylene production processes, and the removal of this contaminant will be considered.
To begin, the formation of acetylene in ethylene product streams will be examined. More than 97% of ethylene around the world is produced by pyrolysis of hydrocarbons, which is the thermal cracking of petrochemicals in the presence of steam. This process can be described as the heating of a mixture of steam and hydrocarbon to the necessary cracking temperature, which can range from 500° F to 650° F depending on the hydrocarbon used. This mixture is then fed to a fired reactor or furnace and heated to between 750° F and 875° F. As a result, the original saturated hydrocarbon “cracks” into smaller unsaturated molecules. This process is extremely endothermic, and the product must be cooled back to the original feed temperature upon leaving the reactor in order to minimize secondary reactions. Possible alkane feedstocks for pyrolysis include ethane, propane, n-butane, isobutane, naphthas, kerosene, and various gas oils. In the U.S., ethane and natural gas liquids (often a mixture of ethane and propane) are most commonly used in ethylene production. Incidentally, using an ethane feedstock produces the smallest amount of acetylene byproduct, which averages about 0.26 wt% of the product stream. For other feeds, this quantity can become as large as 0.95 wt%.
Here is a proposed mechanism for the ethane cracking process:
Initiation
C2H6 → CH3* + CH3*
Propagation
CH3* + C2H6 → C2H4 + C2H5*
C2H5* → C2H4 + H*
H* + C2H6 → H2 + C2H5*
Termination
H* + H* → H2
CH3* + H* → C2H4
C2H5* + CH3* → C3H8
C2H5* + C2H5* → C4H10
etc.
where * denotes a radical.
The reason that acetylene removal from the ethylene stream is so vital is that acetylene acts as a poison to the catalysts used for making polyethylene out of the ethylene product. In addition, acetylene can form metal acetylides, which are explosive contaminants.
Now, methods for purifying the ethylene and removing these trace amounts of acetylene will be examined. After exiting the reactor, the mixed product stream is quenched in order to separate the heavier fractions, and the light and heavy components can be refined individually. The focus should remain on the lighter components at this time. It is vital to remove the acetylene from the cracked gas in order to comply with purity specifications for product ethylene.
Figure 1: A schematic of an ethylene plant.
Source of image: Reference 5, Volume 9 p. 409 (See below)
The overhead stream from the initial separation that was described above is cooled and compressed, and at a later stage there are three main options for removing the acetylene from the mixture of materials. The first and most important method is the selective hydrogenation of the cracked gas in order to convert acetylene to ethylene, the desired product. The other two options are acetylene recovery using a solvent, and fractional distillation. They are outside the scope of this report.
The procedure for hydrogenating the cracked gas will now be examined. The hydrogenation of the cracked gas removes saturated hydrocarbon contaminants present in the product stream. The hydrogenation reaction usually takes place in a packed-bed reactor, and if some olefins are accidentally hydrogenated all the way to saturation, these alkanes can be separated and recycled back to the pyrolytic reactor. The most common placement of this hydrogenation reactor is after the demethanization and deethanization columns as well as after an acid gas removal system. The hydrogenation reaction takes place on a selective catalyst, and contributes to the overall yield of ethylene. Selectivity is a very important aspect of catalyst selection, as undesired reactions will reduce yield and possibly create even more contaminants. Partial hydrogenation of acetylene to ethylene is the goal of this process, and this is possible due to the preferential adsorption of acetylene on the surface of the catalyst when compared to ethylene.
There are several options for catalysts that are feasible for industrial hydrogenation of acetylene in hydrocarbonaceous streams. One commonly used catalyst is alumina-supported palladium, designated Pd/Al2O3. This catalyst is macroporous, so it has a relatively small surface are of approximately 0.1 to 2 m3/gram. It also requires the presence of carbon monoxide to enhance its selectivity. Due to these limitations, a catalyst consisting of silica-supported Pd is preferable. It contains between 0.001 and
1 weight % Pd and between 0.005 and 5% of a promoter metal, preferably potassium. The catalyst is prepared by impregnating the silica support with a solution containing the promoter metal, and then drying the impregnated support. Then this product is impregnated with a solution containing Pd. Finally, the catalyst is dried and calcined. It has a BET surface area of 100 to 300 m3/gram, and is usually processed as an extrudate. It can be activated by reduction with hydrogen or hydrogen-containing gas, which produces a thin layer of Pd over the promoter metal oxide. The operating conditions with this catalyst involve a slight excess of hydrogen, pressure of 10 to 30 bars, and a relatively low temperature, usually less than 150 °C. The reactor operates adiabatically and can reduce acetylene content from the typical value of 0.25 wt % to approximately 1 ppm.
Another catalyst that is available is Pd/Ag/Al2O3. The addition of silver improves the catalyst performance so that only an insignificant amount of ethylene is lost due to overhydrogenation. Like the previous catalyst, it does not require CO. This catalyst is activated by wet reduction with an alkali metal borohydride/alkali metal hydroxide solution, and then dried and heated to enable calcining. It can also be oxidatively regenerated. The catalyst has between 0.01 to 0.2 wt % Pd and 0.02 to 2 wt % Ag, with the preferred Ag:Pd weight ratio being between 5:1 and 8:1. The particles should be spheres or cylindrical pellets with diameters between 2 – 6 mm with a BET surface area of 1 to 100 m3/gram. The feed should be premixed before entering the reactor, and there should be at least one mole of hydrogen for every mole of acetylene in the hydrocarbon stream. This reactor should operate at between 0 – 150 °C and 100 – 1000 psig. It is possible to regenerate the catalyst by heating it up to about 700 °C in the presence of air.
Other possible catalyst options for gaseous feeds are metal oxides and sulfides, usually ZnO, as well as other supported Group VIII metals. Also, if the plant is configured differently it would be possible for the acetylene-bearing hydrocarbon stream to be a liquid. In this case, the proper catalyst would be Pd/Al2O3 with the addition of 0.1 to 1 wt % of an amine compound, and the reactor should be run at 10 – 50 bars and 20 – 150 °C with excess hydrocarbon.
As noted before, some catalysts require an addition of small amounts of carbon monoxide in order to enhance the catalyst selectivity. However, this substance can poison these same catalysts when present at higher percentages, and in addition the carbon monoxide leaves with the hydrocarbon stream. Later, when the stream is processed the carbon monoxide can be harmful to product quality in later reactions and ethylene processing. There is another chemical that could possibly be used in place of CO. Arsine, with a formula of AsH3, can be used as a conversion moderator for acetylene hydrogenation. It is an improvement because it remains on the catalyst and does not leave with the hydrocarbon stream as a contaminant. The purpose of a conversion moderator is to prevent temperature runaway due to the exothermic nature of the hydrogenation process. This prevents the acetylene from being converted all the way to ethane instead of to ethylene.
There is also another option for catalyst regeneration that is especially effective for Pd catalysts, which are the most widespread. The hydrogenation operation results in so-called “green oil” accumulation on the surface of the catalyst. This deactivates the catalyst, so it is necessary to remove it periodically. The most common procedure is to burn off, or oxygenate, the green oil in order to regenerate the catalyst. This must be done every 1 – 3 months. However, this is very time-consuming as it is necessary to purge the reactor of all hydrogen, and it also produces CO and CO2, which are undesirable by-products. In addition, the catalyst must be heated up to a very high temperature, and this shortens the catalyst life due to its heat sensitivity. A viable alternative to this regeneration process is hydrogen stripping. As stated before, this option is especially effective for palladium catalysts as it increases the selectivity of the regenerated catalyst to a level greater than that of fresh catalyst. This may be because it reduces the number of Pd sites responsible for overhydrogenation. Hydrogen stripping involves feeding a mixture of 5 – 10 % hydrogen and the balance nitrogen to the spent catalyst at 350 °C and 50 psig. This is much less time-consuming than oxygenation, since the whole procedure takes between 16 – 24 hours. The only disadvantage to hydrogen stripping is that the catalyst deactivates more quickly so that the procedure must be performed more often.
References:
1.) Brophy, et al. “Selective hydrogenation of acetylene” November 10, 1987.
US Patent # 4,705,906
2.) Cheung et al. “Selective acetylene hydrogenation” April 23, 1996. US Patent # 5,510,550
3.) Cosyns, et al. “Process for selectively hydrogenating acetylene in a mixture of acetylene
and ethylene” February 18, 1986. US Patent # 4,571,442
4.) Flick, et al. “Supported palladium catalyst for selective catalytic hydrogenation of
acetylene in hydrocarbonaceous streams” January 5, 1999.
US Patent # 5,856,262
5.) Hermen F. Mark, Donald F. Othmer, Charles G. Overberger, Glenn T. Seaborg,
Encyclopedia of Chemical Technology, Vols. 1, 9, 15, 18. 3rd. edition, John Wiley & sons, Inc., New York 1978.
6.) Huang, et al. “Regeneration of acetylene converter catalysts by hydrogen stripping”
July 26, 1994. US Patent # 5,332,705
7.) McCue, et al. “Mixed phase front end C.sub.2 acetylene hydrogenation” May 9, 1995.
US Patent # 5,414,170
8.) McFarland, Cecil G. “Acetylene removal process” April 14, 1987. US Patent # 4,658,080
9.) Slim, et al. “Arsine and phosphines as acetylene converter moderators” October 31, 1995.
US Patent # 5,463,154
10.) Ullmann’s Encyclopedia of Industrial Chemistry, Vols. A1, A5, and A10. 5th Edition,
VCH, Weinheim, Germany 1987.
Some informative websites:
LSU Gordon A. and Mary Cain Department
of Chemical Engineering Homepage
The U.S. Patent Office Searchable Database
Dr. Geoffrey
Price's Zeolite Page
Johnson Matthey Chemicals, a source for hydrogenation catalysts
Precious Metals Corporation, specializing in precious metal catalysts
Return to Student Papers
Return to 4205 Homepage
