Antifoulants for Thermal Cracking Processes

This invention relates to processes for the thermal cracking of a gaseous stream containing hydrocarbons. In one aspect this invention relates to a method for reducing the formation of carbon on the cracking tubes in furnaces used for the thermal cracking of a gaseous stream containing hydrocarbons and in any heat exchangers used to cool the effluent flowing from the furnaces. In another aspect this invention relates to particular antifoulants which are useful for reducing the rate of formation of carbon on the walls of such cracking tubes and in such heat exchangers.

The cracking furnace forms the heart of many chemical manufacturing processes. Often, the performance of the cracking furnace will carry the burden of the major profit potential of the entire manufacturing process. Thus, it is extremely desirable to maximize the performance of the cracking furnace.

In a manufacturing process such as the manufacture of ethylene, feed gas such as ethane and/or propane and/or naphtha is fed into the cracking furnace. A diluent fluid such as steam is usually combined with the feed material being provided to the cracking furnace. Within the furnace, the feed stream which has been combined with the diluent fluid is converted to a gaseous mixture which primarily contains hydrogen, methane, ethylene, propylene, butadiene, and small amounts of heavier gases. At the furnace exit this mixture is cooled, which allows removal of most of the heavier gases, and compressed.

The compressed mixture is routed through various distillation columns where the individual components such as ethylene are purified and separated. The separated products, of which ethylene is the major product, then leave the ethylene plant to be used in numerous other processes for the manufacture of a wide variety of secondary products.

The primary function of the cracking furnace is to convert the feed stream to ethylene and/or propylene. A semi-pure carbon which is termed "coke" is formed in the cracking furnace as a result of the furnace cracking operation. Coke is also formed in the heat exchangers used to cool the gaseous mixture flowing from the cracking furnace. Coke formation generally results from a combination of a homogeneous thermal reaction in the gas phase (thermal coking) and a heterogeneous catalytic reaction between the hydrocarbon in the gas phase and the metals in the walls of the cracking tubes or heat exchangers (catalytic coking).

Coke is generally referred to as forming on the metal surfaces of the cracking tubes which are contacted with the feed stream and on the metal surfaces of the heat exchangers which are contacted with the gaseous effluent from the cracking furnace. However, it should be recognized that coke may form on connecting conduits and other metal surfaces which are exposed to hydrocarbons at high temperatures. Thus, the term "Metals" will be used hereinafter to refer to all metal surfaces in a cracking process which are exposed to hydrocarbons and which are subject to coke deposition.

A normal operating procedure for a cracking furnace is to periodically shut down the furnace in order to burn out the deposits of coke. This downtime results in a substantial loss of production. In addition, coke is an excellent thermal insulator. Thus, as coke is deposited, higher furnace temperatures are required to maintain the gas temperature in the cracking zone at a desired level. Such higher temperatures increase fuel consumption and will eventually result in shorter tube life.

Another problem associated with carbon formation is erosion of the Metals, which occurs in two fashions. First, it is well known that in the formation of catalytic coke the metal catalyst particle is removed or displaced from the surface and entrained within the coke. This phenomenon results in extremely rapid metal loss and, ultimately, Metals failure. A second type of erosion is caused by carbon particles that are dislodged from the tube walls and enter the gas stream. The abrasive action of these particles can be particularly severe on the return bends in the furnace tube.

Yet another and more subtle effect of coke formation occurs when coke enters the furnace tube alloy in the form of a solid solution. The carbon then reacts with the chromium in the alloy and chromium carbide precipitates. This phenomena, known as carburization, causes the alloy to lose its original oxidation resistance, thereby becoming susceptible to chemical attack. The mechanical properties of the tube are also adversely effected. Carburization may also occur with respect to iron and nickel in the alloys.

It is thus an object of this invention to provide a method for reducing the formation of coke on the Metals. It is another object of this invention to provide particular antifoulants which are useful for reducing the formation of carbon on the Metals.

In accordance with the present invention, an antifoulant selected from the group consisting of a combination of tin and copper, a combination of copper and antimony, and a combination of tin, antimony and copper is contacted with the Metals either by pretreating the Metals with the antifoulant, adding the antifoulant to the hydrocarbon feedstock flowing to the cracking furnace or both. The use of the antifoulant substantially reduces the formation of coke on the Metals which substantially reduces the adverse consequences which attend such coke formation.

Other objects and advantages of the invention will be apparent from the foregoing brief description of the invention and the claims as well as the detailed description of the drawings in which:

FIG. 1 is a diagrammatic illustration of the test apparatus used to test the antifoulants of the present invention;

FIG. 2 is a graphical illustration of the effect of a combination of tin and copper; and

FIG. 3 is a graphical illustration of the effect of a combination of copper and antimony.

The invention is described in terms of a cracking furnace used in a process for the manufacture of ethylene. However, the applicability of the invention described herein extends to other processes wherein a cracking furnace is utilized to crack a feed material into some desired components and the formation of coke on the walls of the cracking tubes in the cracking furnace or other metal surfaces associated with the cracking process is a problem.

Any suitable organic copper compound may be utilized in the combination of copper and antimony antifoulant, in the combination of tin and copper antifoulant or in the combination of tin, antimony and copper antifoulant. Also, elemental copper and inorganic copper compounds may be used in the combination antifoulants. However, the use of elemental copper and inorganic copper compounds is not preferred since they are not believed to have a significant beneficial effect when used in the combinations.

Examples of organic copper compounds that can be used include complexes of zero-valent copper such as ethylcopper, phenylcopper, 2-tolylcopper, 3-tolylcopper and complexes thereof (e.g., with triphenylphosphine and 2,2'-bipyridine); copper (I) and (II) carboxylates such as formates, acetates, butyrates, octoates, stearates, tallates, oxalates, benzoates and salicylates; copper (II) Retonates or diketonates such as ethylacetonate, acetylacetonate, 3-methylacetylacetonate, 3-ethylacetylacetonate and 3-phenylacetylacetonate; copper (II) alkoxides such as methoxide, ethoxide and phenoxide; copper (II) diethyldithiocarbamate; copper (II) pyridine adduct and the like. Presently preferred are copper (II) tallate and copper (II) 2-ethylhexanoate.

Any suitable form of antimony may be utilized in the combination of copper and antimony antifoulant or in the combination or tin, antimony and copper antifoulant. Elemental antimony, inorganic antimony compounds and organic antimony compounds as well as mixtures of any two or more thereof are suitable sources of antimony. The term "antimony" generally refers to any one of these antimony sources.

Examples of some inorganic antimony compounds which can be used include antimony oxides such as antimony trioxide, antimony tetroxide, and antimony pentoxide; antimony sulfides such as antimony trisulfide and antimony pentasulfide; antimony sulfates such as diantimony trisulfate; antimonic acids such as metaantimonic acid, orthoantimonic acid and pyroantimonic acid; antimony halides such as antimony trifluoride, antimony trichloride, antimony tribromide, antimony triiodide, antimony pentafluoride and antimony pentachloride; antimonyl halides such as antimonyl chloride and antimonyl trichloride. Of the inorganic antimony compounds, those which do not contain halogen are preferred.

Examples of some organic antimony compounds which can be used include antimony carboxylates such as antimony triformate, antimony trioctoate, antimony triacetate, antimony tridodecanoate, antimony trioctadecanoate, antimony tribenzoate, and antimony tris(cyclohexenecarboxylate); antimony thiocarboxylates such as antimony tris(thioacetate), antimony tris(dithioacetate) and antimony tris(dithiopentanoate); antimony thiocarbonates such as antimony tris(O-propyl dithiocarbonate); antimony carbonates such as antimony tris(ethyl carbonates); trihydrocarbylantimony compounds such as triphenylantimony; trihydrocarbylantimony oxides such as triphenylantimony oxide; antimony salts of phenolic compounds such as antimony triphenoxide; antimony salts of thiophenolic compounds such as antimony tris(-thiophenoxide); antimony sulfonates such as antimony tris(benzenesulfonate) and antimony tris(p-toluenesulfonate); antimony carbamates such as antimony tris(diethylcarbamate); antimony thiocarbamates such as antimony tris(dipropyldithiocarbamate), antimony tris(-phenyldithiocarbamate) and antimony tris(butylthiocarbamate); antimony phosphites such as antimony tris(diphenyl phosphite); antimony phospates such as antimony tris(dipropyl phosphate); antimony thio phosphates such as antimony tris(O,O-dipropyl thiophosphate) and antimony tris(O,O-dipropyl dithiophosphate) and the like. At present antimony (III) 2-ethylhexanoate is preferred.

Any suitable form of tin may be utilized in the combination of tin and copper antifoulant or in the combination of tin, antimony and copper antifoulant. Elemental tin, inorganic tin compounds, and organic tin compounds as well as mixtures of any two or more thereof are suitable sources of tin. The term "tin" generally refers to any one of these tin sources.

Examples of some inorganic tin compounds which can be used include tin oxides such as stannous oxide and stannic oxide; tin sulfides such as stannous sulfide and stannic sulfide; tin sulfates such as stannous sulfate and stannic sulfate; stannic acids such as metastannic acid and thiostannic acid; tin halides such as stannous fluoride, stannous chloride, stannous bromide, stannous iodide, stannic fluoride, stannic chloride, stannic bromide and stannic iodide; tin phoshates such as stannic phosphate; tin oxyhalides such as stannous oxychloride and stannic oxychloride; and the like. Of the inorganic tin compounds those which do not contain halogen are preferred as the source of tin.

Examples of some organic tin compounds which can be used include tin carboxylates such as stannous formate, stannous acetate, stannous butyrate, stannous octoate, stannous decanoate, stannous oxalate, stannous benzoate, and stannous cyclohexanecarboxylate; tin thiocarboxylates such as stannous thioacetate and stannous dithioacetate; dihydrocarbyltin bis(hydrocarbyl mercaptoalkanoates) such as dibutyltin bis(isooctyl mercaptoacetate) and dipropyltin bis(butyl mercaptoacetate); tin thiocarbonates such as stannous O-ethyl dithiocarbonate; tin carbonates such as stannous propyl carbonate; tetrahydrocarbyltin compounds such as tetrabutyltin, tetraoctyltin, tetradodecyltin, and tetraphenyltin; dihydrocarbyltin oxides such as dipropyltin oxide, dibutyltin oxide, dioctyltin oxide, and diphenyltin oxide; dihydrocarbyltin bis(hydrocarbyl mercaptide)s such as dibutyltin bis(dodecyl mercaptide); tin salts of phenolic compounds such as stannous thiophenoxide; tin sulfonates such as stannous benzenesulfonate and stannous-p-toluenesulfonate; tin carbamates such as stannous diethylcarbamate; tin thiocarbamates such as stannous propylthiocarbamate and stannous diethyldithiocarbamate; tin phosphites such as stannous diphenyl phosphite; tin phosphates such as stannous dipropyl phosphate; tin thiophosphates such as stannous O,O-dipropyl thiophosphate, stannous O,O-dipropyl dithiophosphate and stannic O,O-dipropyl dithiophosphate, dihydrocarbyltin bis(O,O-dihydrocarbyl thiophosphate)s such as dibutyltin bis(O,O-dipropyl dithiophosphate); and the like. At present tin (II) 2-ethylhexanoate is preferred.

Any of the listed sources of tin may be combined with any of the listed sources of copper to form the combination of tin and copper antifoulant or the combination of tin, antimony and copper antifoulant. In like manner, any of the listed sources of copper may be combined with any of the listed sources of antimony to form the combination of copper and antimony antifoulant or the combination of tin, antimony and copper antifoulant.

Any suitable concentration of copper in the combination of copper and antimony antifoulant may be utilized. A concentration of copper in the range of about 10 mole percent to about 90 percent is presently preferred because the effect of the combination of copper and antimony antifoulant is reduced outside of this range. In like manner, any suitable concentration of copper may be utilized in the combination of copper and tin antifoulant. A concentration of copper in the range of about 10 mole percent to about 90 mole percent is presently preferred because the effect of the combination of copper and tin antifoulant is reduced outside of this range.

Any suitable concentration of anitmony and copper in the combination of tin, antimony and copper antifoulant may be utilized. A concentration of antimony in the range of about 10 mole percent to about 65 mole percent is presently believed to be preferred. In like manner, a concentration of copper in the range of about 10 mole percent to about 65 mole percent is presently believed to be preferred.

In general, the antifoulants of the present invention are effective to reduce the buildup of coke on any of the high temperature steels. Commonly used steels in cracking tubes are Incoloy 800, Inconel 600, HK40, 11/4 chromium-1/2 molybdenum steel, and Type 304 Stainless Steel. The composition of these steels in weight percent is as follows:

The antifoulants of the present invention may be contacted with the Metals either by pretreating the Metals with the antifoulant, adding the antifoulant to the hydrocarbon containing feedstock or preferably both.

If the Metals are to be pretreated, a preferred pretreatment method is to contact the Metals with a solution of the antifoulant. The cracking tubes are preferably flooded with the antifoulant. The antifoulant is allowed to remain in contact with the surface of the cracking tubes for any suitable length of time. A time of at least about one minute is preferred to insure that all of the surface of the cracking tube has been treated. The contact time would typically be about ten minutes or longer in a commercial operation. However, it is not believed that the longer times are of any substantial benefit other than to fully assure an operator that the cracking tube has been treated.

It is typcially necessary to spray or brush the antifoulant solution on the Metals to be treated other than the cracking tubes but flooding can be used if the equipment can be subjected to flooding.

Any suitable solvent may be utilized to prepare the solution of antifoulant. Suitable solvents include water, oxygen-containing organic liquids such as alcohols, ketones and esters and aliphatic and aromatic hydrocarbons and their derivatives. The presently preferred solvents are normal hexane and toluene although kerosene would be a typically used solvent in a commerical operation.

Any suitable concentration of the antifoulant in the solution may be utilized. It is desirable to use a concentration of at least 0.05 molar and concentrations may be 1 molar or higher with the strength of the concentrations being limited by metallurgical and economic considerations. The presently preferred concentration of antifoulant in the solution is in the range of about 0.1 molar to about 0.5 molar.

Solutions of antifoulants can also be applied to the surfaces of the cracking tube by spraying or brushing when the surfaces are accessible but application in this manner has been found to provide less protection against coke deposition than immersion. The cracking tubes can also be treated with finely divided powders of the antifoulants but, again, this method is not considered to be particularly effective.

In addition to pretreating of the Metals with the antifoulant or as an alternate method of contacting the Metals with the antifoulant, any suitable concentration of the antifoulant may be added to the feed stream flowing through the cracking tube. A concentration of antifoulant in the feed stream of at least ten parts per million by weight of the metal(s) contained in the antifoulant based on the weight of the hydrocarbon portion of the feed stream should be used. Presently preferred concentrations of antifoulant metals in the feed stream are in the range of about 20 parts per million to about 100 parts per million based on the weight of the hydrocarbon portion of the feed stream. Higher concentrations of the antifoulant may be added to the feed stream but the effectiveness of the antifoulant does not substantially increase and economic considerations generally preclude the use of higher concentrations.

The antifoulant may be added to the feed stream in any suitable manner. Preferably, the addition of the antifoulant is made under conditions whereby the antifoulant becomes highly dispersed. Preferably, the antifoulant is injected in solution through an orifice under pressure to atomize the solution. The solvents previously discussed may be utilized to form the solutions. The concentration of the antifoulant in the solution should be such as to provide the desired concentration of antifoulant in the feed stream.

Steam is generally utilized as a diluent for the hydrocarbon containing feedstock flowing to the cracking furnace. The steam/hydrocarbon molar ratio is considered to have very little effect on the antifoulants of the present invention.

The cracking furnace may be operated at any suitable temperature and pressure. In the process of steam cracking of light hydrocarbons to ethylene, the temperature of the fluid flowing through the cracking tubes increases during its transit through the tubes and will attain a maximum temperature at the exit of the cracking furnace of about 850.degree. C. The wall temperature of the cracking tubes will be higher and may be substantially higher as an insulating layer of coke accumulates within the tubes. Furnace temperatures of nearly 2000.degree. C. may be employed. Typical pressures for a cracking operation will generally be in the range of about 10 to about 20 psig at the outlet of the cracking tube.

Before referring specifically to the examples which will be utilized to further illustrate the present invention, the laboratory apparatus will be described by referring to FIG. 1 in which a 9 millimeter quartz reactor 11 is illustrated. A part of the quartz reactor 11 is located inside the electric furnace 12. A metal coupon 13 is supported inside the reactor 11 on a two millimeter quartz rod 14 so as to provide only a minimal restriction to the flow of gases through the reactor 11. A hydrocarbon feed stream (ethylene) is provided to the reactor 11 through the combination of conduit means 16 and 17. Air is provided to the reactor 11 through the combination of conduit means 18 and 17.

Nitrogen flowing through conduit means 21 is passed through a heated saturator 22 and is provided through conduit means 24 to the reactor 11. Water is provided to the saturator 22 from the tank 26 through conduit means 27. Conduit means 28 is utilized for pressure equalization.

Steam is generated by saturating the nitrogen carrier gas flowing through the saturator 22. The steam/nitrogen ratio is varied by adjusting the temperature of the electrically heated saturator 22.

The reaction effluent is withdrawn from the reactor 11 through conduit means 31. Provision is made for diverting the reaction effluent to a gas chromatograph as desired for analysis.

In determining the rate of coke deposition on the metal coupon, the quantity of carbon monoxide produced during the cracking process was considered to be proportional to the quantity of coke deposited on the metal coupon. The rationale for this method of evaluating the effectiveness of the antifoulants was the assumption that carbon monoxide was produced from deposited coke by the carbon-steam reaction. Metal coupons examined at the conclusion of cracking runs bore essentially no free carbon which supports the assumption that the coke had been gasified with steam.

The selectivity of the converted ethylene to carbon monoxide was calculated according to equation 1 in which nitrogen was used as an internal standard. ##EQU1## The conversion was calculated according to equation 2. ##EQU2## The CO level for the entire cycle was calculated as a weighted average of all the analyses taken during a cycle according to equation 3. ##EQU3##

The percent selectivity is directly related to the quantity of carbon monoxide in the effluent flowing from the reactor.

Incoloy 800 coupons, 1".times.1/4".times.1/16", were employed in this example. Prior to the application of a coating, each Incoloy 800 coupon was thoroughly cleaned with acetone. Each antifoulant was then applied by immersing the coupon in a minimum of 4 mL of the antifoulant/solvent solution for 1 minute. A new coupon was used for each antifoulant. The coating was then followed by heat treatment in air at 700.degree. C. for 1 minute to decompose the antifoulant to its oxide and to remove any residual solvent. A blank coupon, used for comparisons, was prepared by washing the coupon in acetone and heat treating in air at 700.degree. C. for 1 minute without any coating. The preparation of the various coatings are given below.

0.5M Sb: 2.76 g of Sb(C.sub.8 H.sub.15 O.sub.2).sub.3 (antimony (III) 2-ethylhexanoate) was mixed with enough pure n-hexane to make 10.0 mL of solution referred to hereinafter as solution A.

0.5M Sn: 2.02 g of Sn(C.sub.8 H.sub.15 O.sub.2).sub.2 (tin (II) 2-ethylhexanoate) was dissolved in enough pure n-hexane to make 10.0 mL of solution referred to hereinafter as solution B.

.about.0.1M Cu: Copper(II) 2-ethylhexanoate was prepared by mixing a solution of 42.6 g (0.25 mole) of CuCl.sub.2 -H.sub.2 O in 100 mL of H.sub.2 O with a solution of 83.1 g (0.50 mole) of sodium 2-ethylhexanoate with vigorous stirring. The Cu(C.sub.8 H.sub.15 O.sub.2).sub.2 precipitate was recovered by filtration and extracted with 500 mL of toluene. The extract was evaporated under vacuum conditions so as to recover copper(II) 2-ethylhexanoate crystals. A saturated solution of Cu(.sub.8 H.sub.15 O.sub.2).sub.2 in toluene was prepared which contained about 0.1 mole/1 and is hereinafter referred to as solution C.

0.5M Cu: Copper(II) tallate was prepared by heating basic copper(II) carbonate with tall oil acid in a 100 mL toluene under a nitrogen blanket. The mixture was refluxed until 11.5 mL of H.sub.2 O was distilled off. The cooled mixture was vacuum-filtered through a filter aid-coated filter, and the filtrate was evaporated under vacuum conditions so as to recover Cu(II) tallate. 3.23 g of Cu(II) tallate was dissolved in enough toluene to make 10 mL of a solution referred to hereinafter as solution D.

0.5M Cu-Sb: 1.61 g of copper (II) tallate and 1.37 g of antimony (III) 2-ethylhexanoate were dissolved in enough pure toluene to make 10 mL of a solution referred to hereinafter as solution E.

0.5M Cu-Sn: 1.60 g of copper (II) tallate and 1.03 g of tin (II) 2-ethylhexanoate were dissolved in enough pure toluene to make 10 mL of a solution referred to hereinafter as solution F.

The temperature of the quartz reactor was maintained so that the hottest zone was 900.+-.5.degree. C. A coupon was placed in the reactor while the reactor was at reaction temperature.

A typical run consisted of three 20 hour coking cycles (ethylene, nitrogen and steam), each of which was followed by a 5 minute nitrogen purge and a 50 minute decoking cycle (nitrogen, steam and air). During a coking cycle, a gas mixture consisting of 73 mL per minute ethylene, 145 mL per minute nitrogen and 73 mL per minute steam passed downflow through the reactor. Periodically, snap samples of the reactor effluent were analyzed in a gas chromatograph. The steam/hydrocarbon molar ratio was 1:1.

Table I summarizes results of cyclic runs (with either 2 or 3 cycles) made with Incoloy 800 coupons that had been immersed in the test solutions A-H previously described.

Results of runs 2, 3, 4 and 5 in which tin, antimony and copper were used separately, show that only tin and the organic compound of copper were effective in substantially reducing the rate of carbon deposition on Incoloy 800 under conditions simulating those in an ethane cracking process. Binary combinations of these elements used in runs 6 and 7 show some very surprising effects. Run 6, in which antimony and copper were combined, and run 7, in which tin and copper were combined, show that these combinations were unexpectedly much more effective than results of runs in which they were used separately would lead one to expect. It is also believed that a trinary combination would be more effective than the use of the antifoulants separately.

Using the process conditions of Example 1, a plurality of runs were made using antifoulants which contained different ratios of tin and copper and different ratios of copper and antimony. Each run employed a new Incoloy 800 coupon which had been cleaned and treated as described in Example 1. The antifoulant solutions were prepared as described in Example 1 with the exception that the ratio of the elements was varied. The results of these tests are illustrated in FIGS. 2 and 3.

Referring to FIG. 2, it can be seen that the combination of tin and copper was particularly effective when the concentration of copper ranged from about 10 mole percent to about 90 mole percent. Outside of this range, the effectiveness of the combination of tin and copper was reduced.

Referring now to FIG. 3, it can be seen that the combination of copper and antimony was effective when the concentration of copper was in the range of about 10 mole percent to about 90 mole percent. Again, the effectiveness of the combination of copper and antimony is reduced outside of this range.

Reasonable variations and modifications are possible by those skilled in the art within the scope of the described invention and the appended claims.