US2707702A - Art of coking - Google Patents

Description

May 3, 1955 K. MT wATsoN 2,707,702

ART oF comme Filed Oct. l5, 1949 2 Sheets-Sheet l 6 @Naf/V551? 9 @Naf/vare BY MM 31m/16M A Tram/sys May 3, 1955 K. M, wA-rsoN I2,707,702

ART OF' COKING Filed Oct. 15, 1949 2 Sheets-$heet 2.

TTOFINE/S United States Patent O M ART 0F COKING Kenneth M. Watson, Madison, Wis., assignor to Sinclair Refining Company, New York, N. Y., a corporation of Maine Application October 15, 1949, Serial No. 121,575

4 Claims. (Cl. 202--23) My invention relates to improvements in the art of coking hydrocarbon oils. More particularly, it relates to a novel continuous process in which the hydrocarbon oil is introduced to a bed of coke particles which is maintained in a uidized condition and from which coke product is continuously and selectively Withdrawn.

The art of coking is an old art and represents the efforts of refiners for many years to deal with a serious and recurrent problem inherent in distilling and cracking hydrocarbon oils to produce useful fuels while disposing economically of heavy residual stocks. is severe because of the nature of crude oil and the economic pressure which demands maximum production of light fuels from crude oil. Crude oil is mixture of hydrocarbons, some of which are of high molecular weight and are asphaltic or resemble coke in constitution. ,a

Moreover, cracking hydrocarbon fractions to decompose and rearrange the component molecules into lighter and more desirable molecular aggregations also results in molecular combination under the pyrolytic conditions involved. As a consequence, crude oil distillation and heavy oil cracking operations necessarily produce large amounts of high carbon-residue materials. Coke may be produced as such by direct coking operations or it may be reflected in the high carbon residues of heavy residual fuel oils, tars and asphalts. In any event, the necessity of dealing With coking constituents in hydrocarbon oils represents an economi-c problem of the greatest magnitude for reners. The problem is magnified to the highest degree when crude runs and gasoline demand are at peak levels While the market for heavy residual fuels lags, for then mere disposal of the high carbon-residue materials, as distinct from the operational ditculties, becomes a problemof compelling economic importance.

The economic extremity ofthe coke problem explains how direct coking processes have played such an important role in bothrenery practice and in research i fractions to an ultimate dry coke product while the ulti- 'u mate yield of lighter products including gas and gasoline is increased. Coking processes, however, have long beenl and remain characteristically high cost operations because they are inherently non-continuous and because of the difficult and costly problems involved in coke removal; and handling. Thus heavy oils have been coked at high temperatures in coking chambers with continuous removal of volatile materials overhead, but it has not been possible to do anything but collect the coke along the walls and in the bottom of the chamber until it hasto be shut down for coke removal and cleaning. TheA process of coke removal is diiicult and laborious and requires extensive manual labor or expensive mechanical or hydraulic equipment.

Many continuous coking processes have been proposedbut these have beenl unable either toeliminate substan- The problem 2,707,702 Patented May 3, 1955 tially coke removing and handling costs as a major factor or to avoid coking-up the reactor under practical operating conditions. For example, processes have been proposed which depend upon handling the coke in the form of a slow-moving bed of balls or pellets, but these processes involve a high degree of external recycling of coke in the form of the moving bed requiring a coke lifting system for a relatively small net coke withdrawal. Capital and operational costs are correspondingly high. Continuous processes utilizing suspension techniques have also been proposed but these methods depend upon blowing the coke product from the reaction chamber and generally result in a low degree of reduction to coke and distillate or in reactor coking and shut down. Similarly, processes employing uidized techniques have been impractical because of reactor coking and clogging and because of fluidization diiculties.

I have now found that a continuous coking process can be successfully operated without wall and reactor coking and resultant shut down by maintaining a bed of relatively adsorbent coke particles as seed particles in a iluidized state in the reaction zone if large particles built up by coke laydown are selectively and continuously removed from the reaction zone. To effect selective removal of large coke particles as formed, I use a size classifying or elutriating system opening directly into the fluidized bed of coke particles. The bed comprises seed coke and coke particles formed in the process and is advantageously supported upon a perforated grid plate through which the fluidizing medium is passed. The charge oil is introduced to the reaction zone at one or more points, and I have found that it is advantageous to spray the oil directly into the coke bed in order to minimize the wall coking. The coke formed in the process is selectively deposited on the relatively adsorbent seed coke particles and coke product is selectively withdrawn in the form of the larger particles by means of the internal elutriator which retains the smaller particles in the bed of seed coke and allows the larger yparticles to fall into a transportation and storage system from which they are removed as net coke make. In this way, eiective fluidization is maintained, thus maintaining an equilibrium of the varying particle sizes in the bed, and recirculation of undersize particles as seed coke is obtained within the coke bed itself so that no external recycle is necessary. I have discovered that this procedure efectively overcomes the coke removal problem and eliminates large capital expenditures and operational costs necessitated by conventional expensive and complicated de-coking, handling and de-watering equipment and methods. The coke produced is hard, dry and spheroidal in shape somewhat resembling birdshot or black caviar in appearance. The product is described in detail and claimed in my co-pending application Serial No. 146,143 tiled February 24, 1950 and now abandoned. The lighter components of the charge oil together with the lighter decomposition products and the fluidizing and elutriating media pass overhead to a condensing and recovery system.

The elements of my invention will be more clearly understood by reference to Figure l of the accompanying drawings which represents conventionally and diagrammatically a simplified ilow plan of my process. Black oil feed is charged to the process through line 1i) and is raised to a high temperature of 900 F., for example, in heater 10. Dispersion steam from steam manifold 11, line 12 and superheater 13 is mixed with the feed in line 14 and is introduced directly into the coke bed of reactor 15. at a point slightly above perforated grid plate 1.6.

The grid plate is desirably dish shaped or conical in form in order to promote run down of the larger coke K particles to the opening of the elutriator system. The reactor itself may be swedged er flared upwardly and outwardly in order to promote good fluidization in a bed of substantial length with respect to diameter. Suflicient disengaging space is provided above the coke bed for separation and settling of lighter particles.

Fluidizing steam is introduced to the bottom of reactor 1S from steam manifold 11 through superheater 13 and line 18. Coke draw-off is effected through elutriator 19 which opens directly into the coke bed through the depressed central section of grid plate 16. Elutriating steam is provided from steam manifold 11 and superheater 20 and is introduced near the bottom of elutriator 19 by line 21 and nozzle 22 at a rate suicient to retain small particles within the coke bed as seed coke While permitting the larger particles to fall into lock hopper 23 controlled by seal valve Z4. Lighter' components of thc feed oil and steam pass overhead through line 25 and condenser 26. Heavy oil product may be collected in primary overhead receiver 27 while lighter products and steam pass overhead through line 25.5 and condenser 29 to secondary overhead receiver 30 from which gas may be taken overhead by line 3l. Water is withdrawn as receiver bottoms by line 32 and light oil is taken from above the water layer by means of line 33.

Processing conditions according to my invention are varied somewhat according to the nature of the feed stock and the type of coke and oil products desired. Thus a unit embodying my invention ordinarily charges a black oil such as a heavy reduced crude, a cracking still tar or an asphalt and is run for a maximum yield of clean distillate. Under these conditions I have found that a coke bed temperature of about 900 to l000 F. is desirable. To prevent heater and line coking, the black oil charge is heated to about 875 to 950 F. in a conventional external furnace or heater coil and is sprayed into the reactor bed. Dispersion steam superheated to about the same temperature may be used to improve distribution. Additional heat is put into the bed by superheating the iluidizing and elutriating media to 1000" F. or more.

Reactor temperature, however, may be varied over a broader range, say 800D to 1200 F., in order to obtain higher temperatures to reduce more refractory stocks t0 high ultimate yields. To avoid coking the feed heater and charge lines, direct heating of the coke bed may be advantageously applied. For example, fuel gas may be burned with say 98 per cent theoretical air for injection into the bed at a temperature as high as 3000 to 5000 F. For this purpose, a jet type burner of the type disclosed in application Serial No. 97,142 of Pires et al. tiled June 4, 1949, may be employed. Also a stream of coke may be recirculated through an external heater system in which its temperature may be increased either by indirect heat exchange or by burning a portion of the coke to provide heat.

I have found that the effect of temperature is important with respect to both coke yield based on the feed and rate of coke make in terms of unit bed volume. Coke yield appears to be a function of the vaporization conditions obtained within the reaction zone and thus the temperature and partial pressure of the oil. Coke yield appears to decrease with an increase in temperature and a decrease in partial pressure. Rate of coke make per unit bed volume, however, appears te be a function of the factors controlling cle-agglomeration of the coke particles within the bed, particularly the rate of reaction of liquid adhesive material to hard dry coke. Rate of coke make therefore appears to be substantially increased with increase in reactor temperature. 'f he relationship between the coke yield on the feed and the rate of coke make in terms of unit bed volume thus determines the volume of the coke bed for a given feed stock under desired conditions of temperature and pressure.

The reactor pressure is advantageously kept relatively 4f low, ranging from about atmospheric to about 75 p. s. i. Where cleaner' overhead is desired, higher pressures are advantageously employed with concomitant increase in coke yield.

The density of the liuidized coke bed varies from about 25 to about 50 pounds per cubic foot but is generally of the order of about 35 to 40 pounds per cubic foot. I have found that the size range of the seed coke for start-up should be relatively narrow for ease in lluidization. Thus I have obtained satisfactory results with l0 to 30 mesh particles, but I have found that coke ranging in size from as large as about 1A: inch to below l0 microns was difficult to fluidize and tended to result in channelling. In steady state operation, however, the size of the coke particles in the reactor will reach an equilibrium distribution determined by the overall process conditions.

I have found that a iuidizing steam flow of as low as about 2 to 3 feet per second is sufficient for l0 to 30 mesh coke particles. Higher superficial steam velocities produce improved iiuidization but increase steam consumption and of course ultimately are limited by excessive carry-over if sufficient disengaging space above the coke bed is not provided. In a typical operation, I have found that about 35 feet per second superficial velocity for elutriator steam is desirable to yield a product having a mesh range of 10-20. For continuous operation, the ow of the elutriating medium is adjusted according to the rate of coke make, and the size range of the coke with drawn is determined by the process conditions as described above.

Although I consider that steam is peculiarly adapted for use as a fluidizing and elutriating medium in view of its condensability which prevents compressor and condenser overloading and in view of its high heat capacity, other media may be used. For example, flue gases having a low oxygen content or an inert gas such as nitrogen or a naphtha or light refractory gas oil may be utilized. The use of a heavy naphtha or a relatively refractory light cycle stock such as the effluent from a cracking coil heater has significant advantages in itself in cracking operations. Processes representing this application of the principles underlying my invention are disclosed and claimed in my copending application Serial No. 146,146 led February 24, 1950.

The principles underlying my invention will be further illustrated in the following example representative of a series of pilot plant runs. A swedged reactor was used having a diameter of 6 inches at the bottom. At a point 4 inches above the grid, the reactor gradually swedged out to 10-inch diameter at the rate of 1 inch increase in diameter for a vertical rise of 3 inches. A high velocity region in the bottom of the reactor where poor fluidization is most apt to occur is thus made possible with a lower velocity at the top of the reactor where coke fines may be carried overhead. The grid used was conicalor funnelshaped, high at the Walls of the vessel and sloping downward toward the center, so that large coke particles settling to the bottom of the bed would tend to slide or be funneled downward along the grid toward the center where the elutriator mouth was located. Smaller particles were kept from entering the elutriator standpipe by the high velocity elutriating vapor stream which was adjusted to allow the larger particles to be removed from the system. The elutriator was 11/2 inches in diameter, and the grid was 6 inches in diameter sloping at a 30 angle with the horizontal. To obtain even distribution of the fluidizing medium the conical grid was designed with larger holes near the center and smaller holes near the outside in order to overcome the difference in pressure drop across the bed due to the slope of the grid.

In starting the unit a charge of 2l pounds of electrode coke sized from l0 to 30 mesh was placed in the reactor as seed for the formation of process coke. The feed oil was mixed with steam and brought up through the bottom of the reactor and sprayed through a 1A inch hole directed upward at a center point some 3 inches above the bottom of the coke bed in order to minimize wall coking. The feed stock represented a bottoms fraction from the gas oil evaporator of a typical commercial crude topping and thermal cracking combination unit. The stock had a gravity of 7 .7 API, viscosity of 24 SFS at 122 F., sulfur of 1.33%, extraction sediment of 0.04 and a carbon residue of 13.6 (Conradson).

Three independent steam ows were introduced to thel unit. Elutriator steam was introduced at 1050 F. at the bottom of the elutriator standpipe; fluidizing steam entered the reaction zone through the perforated grid to provide the necessary liuidiz'ing medium at about 925 F.; and dispersion steam was introduced with the black oil feed at about 900 F. to provide a spraying action for the oil. The reactor bed temperature was controlled at 911 to 925 F. while the black oil and dispersion steam charge temperature varied from 875 to 915 F. The reactor pressure was maintained at 3 to 4 p. s. i. g. The coke bed had a tluidized density varying between about 33 to 45 pounds per cubic foot and a bed volume of 0.6 to 0.8 cubic foot.

The black oil feed was charged at rates varying from 2 to 8 gallons per hour while the dispersion steam ow was varied up to 10 pounds per hour. Fluidizing steam flow ranged from to 75 pounds per hour and the eluatriator steam flow from 10 to 30 pounds per hour. The coke produced was hard, dry, spherical and 6 to 10 mesh in size. The volatile combustible content was 2 to 8 per cent and the sulfur content less than 1 weight per cent. The yield averaged 3 to 7 weight per cent on the fresh feed, amounting to a coke laydown of 1.5 to 2.2 pounds per hour per cubic foot of bed volume.

The runs were all characterized by the absence of coke build up on the walls of the reactor vessel. Apparently the adsorbency of the coke bed was suiicient to prevent any material with a coking tendency from leaving the bed. Although the ratio of oil to steam was kept low in these runs, it may be greatly increased for a large diameter reactor typical of a commercial unit where the bed height can be safely raised while maintaining the same iluidization. height for a constant cross sectional area while steam rate remains constant, the oil/ steam ratio will increase greatly with increase in the size of a unit. Also because of the high steam quantities, yields were relatively low. Rates of coke make were also somewhat low because of the relatively low temperature. At higher temperatures, I have obtained higher rates; e. g. about 6 pounds per hour per cubic foot of bed volume at 975 F. reactor temf perature.

In commercial operation, of course, my process ordinarily will be combined with related refinery crude running or cracking operations to obtain maximum conservation of heat and to minimize rehandling of stocks. example of a typical tie-in with a topped crude distillation unit is illustrated in Figure 2 of the accompanying drawing. The feed stock representing for example a 20 per cent virgin reduced crude is charged to the unit through line 100. The charge is advantageously heated by exchange with hot products in the usual manner and is introduced to fractionating tower 101 at a point in the lower section. Overhead from the tower is removed through line 102 and condenser 103 to receiving drum 104. A

light gas fraction is removed from receiving drum 104 by t line 105, water is withdrawn from the bottom of drum 104 by line 106 and condensate is removed by line 106, of which part is taken as product through line 107 and part is returned as reflux to the tower through line 108. One

or more side streams are advantageously drawn, and as shown a light gas oil stream is taken from an upper portion of tower 101 through line 109 to receiving drum 110 equipped with vapor return line 111 and product draw-off line 112. Similarly, a heavy gas oil cut is taken from a lower portion of the tower through line 113 and receiving Since the permissible oil feed increases with bed` to a lower section of fractionating tower 101.

6 drum 114, equipped with vapor return line 115 and product draw-oi line 116.

Bottoms from fractionating tower 101 represents the feed to coking reactor 117 and is charged through line 118 and heater 119. Dispersion steam may be mixed with the feed through line 120 and the feed is introduced advantageously into the coke bed at a point just above perforated conical grid plate 121. A bottoms stream may also be used to generate steam by passage through line 122 and steam boiler 123, being returned by line 124 to fractionatin g tower 101. Water is supplied to steam boiler 123 through line 125 and the steam produced is passed by line 126 through steam superheater 127. Make-upA steam may be added or excess withdrawn through line 126. Fluidizing steam is thus provided through line 128 for introduction below the perforated grid at the bottom of coking reactor 117, and elutriating steam is provided for injection into the bottom of elutriator 129. Overhead products from coking reactor 117 are passed by line 130 Vapors leaving coking reactor 117 may be passed through a nest of cyclones for removal of fine coke particles carried overhead.

Elutriator steam flow to elutriator 129 is controlled so as to remove large coke particles as formed by permitting them to drop into coke receiving hopper 132. Hot coke from hopper 132 is passed by line 135 under regulation of slide valve 134 by means of a carrier medium such as water, steam or ilue gas introduced at 136, to hot coke hopper 133 which is provided for cooling and storing coke product, For shut-down or start-up the unit hopper 133 is designed to receive and hold the charge of coke from coking reactor 117. Water spray 137 may be provided in hopper 133 for cooling purposes.

A unit of this type, for example, may charge 15,000 barrels per day of fresh feed such as a 10 API-gravity, 20 per cent virgin reduced crude, and employ a coking reactor of 15 x 80 including a 20-foot disengaging zone. The steam flow amounts to about 40,000 pounds per hour, and the coke draw-off amounts to about 400 tons per day. The coking reactor is operated at about 850 to 910 F. and about 10 to 30 p. s. i. g., while the functionating tower is operated conventionally.

My process provides a continuous method for coking heavy hydrocarbon oils without costly de-coking and coke handling costs. It provides a system for laying down coke selectively in a bed of seed coke particles maintained in a fluidized state while large particles formed in the process are selectively removed by means of an internal elutriator. Obviously, much of the equipment and some of the process steps and conditions are conventional and can be changed in form and degree without aiecting the principles of my invention.

I claim:

l. A method for coking hydrocarbon oils which comprises maintaining in a reaction zone and at a temperature of the order of about 800 F. to 1200 F. a densephase fluidized bed of coke particles comprising coke particles of various sizes, introducing a hydrocarbon oil to the bed of coke particles, introducing a fluidizing medium to the bed of coke particles at a rate such that the velocity of the vapors rising through the bed of coke particles is such as to maintain the bed of coke particles in a dense-phase iuidized state, removing vaporized materials from the reaction zone, withdrawing from the coke bed substantially only the large coke particles in an amount suicient to maintain an equilibrium of the varying particle sizes in the bed to maintain effective iluidization and permitting the withdrawn particles to settle through an elutriating zone countercurrently to a rising stream of elutriating medium owing at a velocity greater than the velocity of the vapors rising through the bed 0f coke particles but not exceeding that at which the large coke particles will settle through the elutriating zone,

7 and removing from the elutriating Zone the large particles which have settled therethrough.

2. A process according to claim 1 in which the large coke particles settling to the bottom of the coke bed are funneled to the elutriating zone.

3. A method for coking hydrocarbon oils which comprises maintaining a dense phase bed of coke particles in a reaction zone at a temperature of the order of about 800 to 1200 F. and in a uidized state, introducing a hydrocarbon oil to the coke bed, selectively withdrawing large coke particles from the coke bed in an amount su'icient to maintain an equilibrium of the varying particle sizes in the bed to maintain effective iluidization by passing an elutriating medium into the coke bed at a velocity which is greater than the velocity of the vapors rising through the coke bed but not exceeding that at which the large coke particles will settle out of the coke bed thereby retaining relatively small colte particles in the coke bed and permitting the separation of the large coke particles therefrom, and removing vaporized materials from the reaction zone.

4. A method for coking hydrocarbon oils which comprises maintaining a dense-phase bed of coke particles in a reaction Zone at a temperature of the order of about 800 to 1200 F. and in a uidized state, introducing a hydrocarbon oil in the coke bed, tunneling large particles settling in the coke bed to a coke draw oi point, selectively withdrawing large coke particles from the coke bed in an amount sutticient to maintain an equilibrium of the varying particle sizes in the bed to maintain effective uidization by passing an elutriating medium into the coke bed at said draw ot point at a velocity which is greater than the velocity of the vapors rising through the coke bed but not exceeding that at which the large Coke particles will settle out of the coke bed thereby retaining relatively small coke particles in the coke bed and permitting the separation of the large coke particles therefrom, and removing vaporized materials from the reaction zone.

References Cited in the file of this patent UNITED STATES PATENTS 2,340,974 Myers Feb. 8, 1944 2,362,270 Hemminger Nov. 7, 1944 2,366,055 Rollman Dec. 26, 1944 2,426,848 Tuttle Sept. 2, 1947 2,445,328 Keith July 20, 1948 2,456,796 Schutte Dec. 2l, 1948 2,485,315 Rex et al. Oct. 18, 1949 2,511,088 Whaley June 13, 1950 2,534,728 Nelson et al. Dec. 19, 1950 2,543,884 Weikart Mar. 6, 1951 2,608,526 Rex Aug. 26, 1952 FOREIGN PATENTS 612,815 Great Britain Nov. 18, 1948

Claims (2)

1. A METHOD FOR COOKING HYDROCARBON OILS WHICH COMPRISES MAINTAINING IN A REACTION ZONE AND AT A TEMPERATURE OF THE ORDER OF ABOUT 800* F. TO 1200*F. A DENSEPHASE FLUIDIZED BED OF COKE PARTICLES COMPRISING COKE PARTICLES OF VARIOUS SIZES, INTRODUCING A HYDROCARBON OIL TO THE BED OF COKE PARTICLES, INTRODUCING A FLUIDIZING MEDIUM TO THE BED OF COKE PARTICLES AT A RATE SUCH THAT THE VELOCITY OF THE VAPORS RISING THROUGH THE BED OF THE COKE PARTICLES IS SUCH AS TO MAINTAIN THE BED OF COKE PARTICLES IN A DENSE-PHASE FLUIDIZED STATE, REMOVING VAPORIZED MATERIALS FROM THE REACTION ZONE, WITHDRAWING FROM THE COKE BED SUBSTANTIALLY ONLY THE LARGE PARTICLES IN AN AMOUNT SUFFICIENT TO MAINTAIN AN EQUILIBRIUM OF THE VARYING PARTICLE SIZES IN THE BED TO MAINTAIN EFFECTIVE FLUIDIZATION AND PERMITTING THE WITHDRAWAL PARTICLES TO SETTLE THROUGH THE ELUTRIATING MEDIUM FLOWING AT A VELOCITY GREATER STREAM OF ELUTRIATING MEDIUM FLOWING AT A VELOCITY GREATER THAN THE VELOCITY OF THE VAPORS RISING THROUGH THE BED OF COKE PARTICLES BUT NOT EXCEEDING THAT AT WHICH THE LARGE COKE PARTICLES WILL SETTLE THROUGH THE ELUTRIATING ZONE, AND REMOVING FROM THE ELUTRIATING ZONE THE LARGE PARTICLES WHICH HAVE SETTLE THERETHROUGH.

2. A PROCESS ACCORDING TO CLAIM 1 IN WHICH THE LARGE COKE PARTICLES SETTLING TO THE BOTTOM OF THE COKE BED ARE FUNNELED TO THE ELUTRIATING ZONE.

Images