US 3155608 A
Description (OCR text may contain errors)
Nov. 3, 1964 J R. HOPPER Filed Dec. 1, 1960 GAS TO GAS 0|| RECOVERY SYSTEM CHARGE FRESH H2 1 STAGE J2ND.STAGE REACTOR REACTOR SEPARATQ COQOLER SEPARATOR 4a 44 CATALYTIC CRACKING UNIT INVENTORS.
JACK R. HOPPER GEORGE P. REYNOLDS,JR., WILLIAM B. FRANKLIN,
United States Patent 3,155,608 PROCESS FOR REDUCING METALS CGNTENT 0F (IATALYTE CRACK- 1N G FEEDSTQCK Jack R. Hopper, llaytown, George P. Reynolds, din, Houston, and William B. Franklin, Baytown, Tex, assignors, by mesne assignments, to Esso Research and Engineering (Iompany, Elizabeth, N..l., a corporation of Delaware Filed Dec. 1, 1960, Ser. No. 72,998 1 Claim. (Cl. 298251) The present process deals with the preparation of a feed stock for catalytic treatment of a petroleum hydrocarbon liquid. More particularly, it deals with the hydroclesulfurization of a feed stock for a catalytic cracking process. In its most specific aspect, the present invention deals with the preparation by hydrodesulfurization of a catalytic cracking feed stock which is substantially sulfur free and is low in contaminating metals content and aromatic compounds.
The catalytic cracking of petrolernum hydrocarbons is a accomplished by the use of a catalyst either in a fixed bed or in the fluidized state. The reaction takes place in the vapor phase, at temperatures in the range of about 900 F. to about 1000 F. A silica-alumina catalyst which is sensitive to sulfur and to certain metallic elements is normally used. Contamination of the catalyst in the catalytic cracking operation shifts the reaction toward the production of normally gaseous hydrocarbons which are unsuitable as motor fuel; toward the production of unsaturated hydrocarbons which are unstable in storage and are less suitable for motor fuel in some instances than the saturated hydrocarbons; and in the production of coke which represents a net loss of feed stock, as well as increasing the heat load on the regenerator of the catalytic cracking unit and the cooling system for the regenerator. These ill results accrue when the feed stock to the catalytic cracking unit contains contaminants which have been referred to above. It is important, therefore, to reduce the contaminant content of the feed stock to a practical minimum. It should be noted that the effects of metal contamination of the catalytic cracking catalyst is cumulative; that is, the metals will not be burned oil and removed from the catalyst in regeneration, and are kept at an acceptable level in the normal practice of the catalytic cracking process usually by continuously removing a portion of the catalyst and replacing it with fresh, uncontaminated cracking catalyst. It is apparent that this is a costly expedient and that by reducing the amount of metal contaminants introduced with the feed stock, the amount of catalyst withdrawal can be reduced. An increase in the selectivity of the catalyst which remains in the cracking unit inventory is also provided.
By the practice of the present invention, the feed stock to a catalytic cracking unit is treated to remove the sulfur compounds therein by splitting off the sulfur from the organic molecule and forming a gaseous hydrogen sulfide (which is easily separated from the normally liquid hydrocarbon product) and adding hydrogen to the molecule to replace the sulfur which has been removed. It is to be understood that the sulfur compounds normally present are thiophenic, but other sulfur compounds are also removed by hydrodesulfurization, as is well known in the art. Further, by the practice of the present invention, the amount of metal contaminants present in the hydrocarbon feed stock are reduced. These metallic contaminants normally are present as organo-metallic compounds which are soluble in the hydrocarbon stream, and ordinarily are not removed by sedimentation or filtration. The metals which contaminate the catalytic cracking catalyst and which are partially removed by the practice of the present invention comprise nickel, vanadium, and iron. Still further, by the alsasas Patented Nov. 3, 1964 practice of the present invention, a significant portion of the heavy aromatic compounds in the oil are hydrogenated either completely or partially to naphthenic compounds.
The practice of the present invention can be more particularly understood by reference to the drawing wherein a preferred mode of the present invention is disclosed.
By the practice of the present invention, the hydrodesulfurization is accomplished in two reactors: a first-stage reactor 10 and a second-stage reactor 12. The fresh charge is introduced into the system by way of line 14. A hydrogen-rich gas is mixed with the charge by means of line 16 from a source later to be discussed. The combined charge and hydrogen-rich gas are introduced into the first-stage reactor 10 by means of line 13. In the firststage reactor, hydrodesulfurization is accomplished under relatively severe conditions and the hydrodesulfurized product is discharged through line 20 and is cooled in cooler 22 from whence it is dicharged by way of line 24 into a separator 26. Within the separator 26 the light gases which have been carried from the first-stage desulfurization are discharged overhead through line 28. These gases comprise light hydrocarbons and hydrogen sulfide formed in the reactor, as well as the unreacted hydrogen and light hydrocarbons which are introduced to the reactor through line 16. The liquid product is discharged from separator 26 through line 30 and is pumped by means 32 via line 34 to the second-stage reactor 12. Before introduction into the second-stage reactor, the liquid product of the first-stage hydrodesulfurization is mixed with a fresh hydrogen gas which is introduced in the heated state by way of line 36. The mixed stream is introduced into reactor 12 by means of line 38. In the second-stage reactor, the reaction is carried on at relatively mild conditions and the product is withdrawn by way of line 40 and introduced into a separator 42 wherein the liquid product is separated and removed from the system as a gas-oil product which is substantially sulfur free and which has had the metallic contaminants and aromatic ring compounds substantially reduced. At least a portion of the gaseous phase which is discharged from separator 42 is carried by way of line 16, as previously discussed, for admixture with the fresh charge. If desired, a part of the gaseous phase may be discharged from the system via line 44 controlled by valve 46. The liquid phase discharged from separator 42 is carried by line 48 to a catalytic cracking unit 50, which may be of a fluidized bed or fixed bed type. 7
Alternatively to the method described above wherein the liquid is pumped and the gas flows by pressure (which is the preferred method), under certain conditions it may be desirable to let the first-stage liquid flow by pressure from drum 26 to reactor 12 and the gas from drum 42 would be compressed to flow to the first-stage reactor 10. For such a flow setup the second-stage reactor would operate at 50 to 150 p.s.i. lower pressure than the first-stage reactor.
Referring now to the first-stage reactor wherein the charge is admixed with the hydrogen-rich gaseous product of separator 42, the severe hydrodesulfurization conditions comprise a temperature within the range of 730 F. to 800 F. at a pressure of 650 to 3000 p.s.i.g. The space velocity utilized in the first-stage reactor is between 1.0 and 4.5 volumes of charge per volume of catalyst per hour. The hydrogen-treat rate in this first-stage reactor suitably is between 350 and 1500 standard cubic feet per barrel of charge with the hydrogen concentration in the feed gas normally being between 65% and by volume, but may be as low as 60%. The hydrogen consumed in this operation ranges between and 250 standard cubic feet of hydrogen per barrel of gas-oil charge. The sulfide concentration in the feed stock at this stage normally is between /2% and 3% by weight.
In the second-stage reactor, a relatively mild hydrodesulfurization is accomplished at a temperature ranging between 600 F. and 725 P. which is approximately 50 F. to 150 F. lower than the temperature in the first stage. The pressure in the second stage is about 70 to 125 p.s.i.g. higher than that used in the first stage, ranging bet-ween 750 and 3000 p.s.i.g. The space velocity in the secondstage reactor suitably is identical to that in the first stage. The hydrogen-treat rate in the second-stage reactor may be between 400 and 1500 standard cubic feet per barrel of charge at a hydrogen concentration about 5% to 15% tigher than in the first stage, preferably in the range of 70% to 90%. The hydrogen consumed in the second stage may range between 20 and 120 standard cubic feet per barrel of charge. The lesser amount of hydrogen here consumed is a result of the prior treatment in the first stage wherein the bulk of the hydrodesulfurization takes place. The amount of sulfide present in the feed stock to the second-stage reactor is between 0% and 2% by weight, depending on the desulfurization accomplished in the first stage.
As compared to the process of the present invention wherein the reactors are placed in series in the process system, with the fresh hydrogen introduced into the second stage and conducted therefrom into the first stage, the prior art has utilized a parallel or once-through hydrodesulfuriz-a-tion wherein the admixture of hydrogen and hydrocarbon charge is passed through the reactors in one pass. As shown below, the present invention Ofi ers a substantial improvement over this once-through hydrodesulfurization.
In order to evaluate the present invention, a series of runs were performed with data being taken in yield periods of hours each. The series of runs comprised three evaluations of the present process wherein only the temperatures used in the first and second stages were varied, and one evaluation of the once-through process of the prior art. The oil rate or space velocity was maintained at substantially the same level during each series run at about twice the space velocity in the parallel run, to give equivalent reaction times. The pressure was maintained the same; and the hydrogen rate and H 8 rate were maintained substantially the same. The operating conditions are tabulated below in Table I:
The feed stock used in the above experiments was prepared by blending process gas oil and deasphalted oil to provide a mixture which is representative of the type of feed stock normally encountered in this type of process.
The "feed stock had an average boiling point of 865 F., a gravity of 22.4 API, and a sulfur content of 1.78% by weight. The metals content of the charge was 0.65 ppm. nickel, 0.88 ppm. vanadium, and 0.55 ppm. iron. The aromatic ring content was about 12.3 weight percent. The hydrogen input into the second stage was about 98+% pure and was introduced into the unit at 800 psig The sulfur content above referred to was induced in the feed stock by metering hydrogen sulfide into the gas-oil charge stream, in order to simulate plant conditions.
The catalyst used in these experiments was employed in the form of A3" extruded pellets. The catalyst is cobalt molybdate upon an inert carrier, having a composition T able II SULFUR LEVEL OF FINAL PRODUCT (WT. PERCENT) Yield Percent Period Feed 1 2 3 4 Avg. Desulturization Parallel 1.78 Series:
A B. O
CONRADSON CARBON CONTENT (WT. PERCENT) Yield Percent Period 1 Feed 1 2 3 4 Avg. Re-
duction Parallel 1. 63 .84 .68 79 .83 79 51. 5 Series:
A 65 60.1 B .62 C .68 58.3
NITROGEN CONTENT (WT. PERCENT) Yield Percent Period 1 Feed 1 2 3 4 Avg. Re-
duction Parallel .14 Series:
A B O NICKEL CONTENT (PARTS PER MILLION) Yield Percent Period 1 Feed 1 2 3 4 Avg. e;
duction Parallel 65 21 14 14 .08 14 78. 5 Series: A 24 08 87. 7 B 24 24 C 10 105 83. 8
AROBIATIC RING CONTENT (WT. PERCENT) 1 For Series: Denotes yield periods of second pass.
2 Composite of yield period 1 and 2.
Readings alter the first yield period in Run 13 were not obtained.
As can be seen by the material tabulated above, the percent desulfurization obtainable by the practice of the present invention is above in the two completed runs shown (i.e., in Series A and Series C), amounting to 92.1% desulfurization at the lowest. The amount of carhon-forming compounds which have been reduced during the hydrodesulfurization is also shown to be quite good, amounting to a minimum of 58.3%. The reduction of nitrogen has been shown to be substantial, within the range of 29.3% to 42.1% reduction.
The metallic contaminants in the cracking operation consist of nickel, vanadium and iron. Since it was known that vanadium and iron are reduced in a relative proportion to the amount of nickel removed, the nickel analysis is representative of the other metal analyses. Hence, the selection of the best operation for reduction of metal con tarninants is based on the reduction of nickel. It should be noted that in the A series operation, 87.7% of the nickel has been removed whereas in the C series operation, 83.8% of the nickel has been removed, as compared to 78.5% reduction in the parallel reactors. In this regard, it should be noted that the analysis of .24 ppm. in the first yield period of the A series operation was disregarded in the average because the time allowed before taking the reading may have been insufiicient to obtain good distribution of the oil on the catalyst.
The reduction of aromatic ring compounds in the feed stock has also been shown to be quite good, amounting to between 20.1% and 26.9%. The advantage shown in the C series operation in the reduction of these aromatic ring compounds is thought to be due to two reasons. First, the rate limitations had been decreased because the sulfur removal in the first pass had decreased the competition for catalyst surface in the second pass. Secondly, since the rate limitations have been decreased, greater conversions are possible because the lower temperature used in the C series favors the equilibrium in hydrogenating the polycyclic aromatics.
In comparing the percent de sul furization attained by the i use of the present invent-ion with that attained by the parallel method previously used, it is seen that a desulfurization level of above 92% has been attained as compared with 85% in the parallel system. The percent reduction of carbon-forming compounds 'by the practice of the present invention has been shown to be between 58% and 60% as compared with around 52% by the prior method. The nitrogen content reduction by the practice of the present invention has been shown to be comparable, at least, to that experienced by the prior method. In the reduction of metallic contaminants, the percent reduction by the practice of the present invention has been shown to be as much as 87.7% as compared to 78.5 obtained by the practice of the parallel hydrodesulturization. The taromati'c ring content of the feed stock, likewise, has bwn shown to be reduced in greater measure by the practice of the present invention as compared to the prior art parallel system of hydrodesulturization. Therefore, it is obvious that the practice of the present invention offers a substantial benefit as compared to the parallel operation previously used.
While not to be limited by the following explanation, it is postulated that the unexpected and unobvious advantage is due to the employment of higher temperatures n the first-stage hydrodesulifurizer which favor the removal of sulfur compounds by hydrogenolysis. Secondly, the removal of the majority or hydrogen sulfide and light bydrocarbons between stages (i.e., in separator 26) raises the hydrogen partial pressure and lowers the hydrogen sulfide partial pressure in the second stage reactor, thus improving the equilibrium conversion of sulfur, nitrogen, condensed ring aromatics and other deleterious compounds in the oil. Thirdly, by lowering the temperature in the second-stage hydrodesuhfurization, the equilibrium for the hydrogenation reaction is t'avored. This condition particularly favors the removal of the aromatic ring cornpounds. Fourthly, and quite significantly, the introduction of fresh hydro-gen to the second stage reaction, wherein the milder conditions prevail, gives a higher hydrogen partial pressure at the point in the process where the concentration of non-hydrogenated compounds has been reduced; that is, because of the previous treatment in the first-stage hydrodesul furization reactor. This results in a more complete conversion of the charge which has already been partially hydrodesulfiurized than could otherwise be obtained.
The feed stock which may be suitably treated by the present process is not limited to the specific blend stocks disclosed above, but generally is a hydrocarbon stream having an API gravity within the range of about 9 to 30 API, and boiling within the range of 400 F. to 1300 F. A suitable feed stock could be, for example, a mixture of 6 product gas oil (PGO) boiling within the range of 400 F. to 1050 F. and a deasphalted oil (DAO, produced, for example, by propane extraction of crude residua) to produce a hydrocarbon stream boiling within the range of 750 F. to 1300 F. The vacuum flashed crude residuum can also be washed with light catalytic cycle oil prior to hydrodesulfurization to lower the metal contamination thereof. The mixture of gas oil to deasphalted oil may be in a ratio within the range of 1.521 to 5:1. Boiling ranges of the product gas oil and deasphalted oil, along with two exemplary feed stock blends of the two, is given below:
Table III Percent PGO DAO 65% PGO- 74% PGO 35% DAO 26% DAO The catalysts used in the practice of the present invention can be any of the Well-known hydrodesulfurization catalysts besides the cobalt molybdate catalysts mentioned herein. These catalysts include nickel-tungsten sulfide, molybdenum sulfide, and nickel sulfide. It should be understood that the hydrogen purity ranges stated herein are not critical and, further, that the purity of the hydrogen stream introduced into the first-stage reactor may suitably be controlled by the conditions in the separator 42 as well as being affected by the purity of the fresh hydrogen stream which is introduced into the second-stage reactor 12. A suitable source of hydrogen for introduction into the second-stage hydrodesulfurization reactor is found in the net hydrogen-make stream of a catalytic reformer. This efiiuent gas may suitably be composed as tabulated below:
Table IV Vol. percent Hydrogen 89.0 Methane 2.7 Ethane 2.5 Propane 2.6 Butane 1.9 Pentane 0.7 Hexane and above 0.6
Other sources of hydrogen of comparable purity and suitable volume may be satisfactorily used.
Applicants having disclosed in detail the basic invention herein involved as well as a preferred and best mode of practicing the same, what is to be protected by Letters Patent should be understood to be limited only by the appended claim and not by the specific examples hereinbefore given.
In the process of catalytically cracking, at a temperature of 900 F. to 1000 F. and in contact with a cracking catalyst which is sensitive to nickel, vanadium, and iron, a hydrocarbon stream boiling within the range of 400 F. to 1300 F., having a gravity of 9 API to 30 API and containing about 0.65 ppm. nickel, and vanadium and iron contaminants,
the improvement of treating said hydrocarbon stream to remove at least a portion of said contaminants by hydrotreating the admixture in a first-stage reactor in the presence of a cobalt molybdate catalyst at a temperature within the range from about 730 F. to
about 800 F and a pressure within the range from about 650 p.s.i.g. to about 3000 p.s.i.g.,
a hydrogen charge rate of 350 to 1500 s.c.f./bbl. of
and a space velocity of 1.0 to 4.5 volumes of hydrocarbon per hour per volume of catalyst,
separating the liquid product of the first-stage reactor from the gaseous eflluent thereof,
hydrotrcating said liquid product in a second-stage reactor in the presence of a cobalt molybdate catalyst at conditions corresponding to a temperature which is from 50 F. to 150 F. lower than the temperature utilized in the first-stage reactor, and at a pressure which is from 70 p.s.i. to 125 psi. higher than that utilized in the first-stage reactor,
at a hydrogen charge rate of 400 to 1500 s.c.f./bbl. of
hydrocarbon feed into the second-stage reactor,
and at a space velocity of from 1.0 to 4.5 volumes of hydrocarbon per hour per volume of catalyst,
separating the effluent of the second-stage reactor into a gaseous portion and a liquid portion,
introducing the gaseous portion into admixture with the hydrocarbon charge stream as said hydrogenrich gaseous stream,
whereby a charge stock for catalytic cracking is obtained which has a substantially reduced metals content.
References Cited in the file of this patent UNITED STATES PATENTS 2,760,907 Attane et a1. Aug. 28, 1956 2,769,754 Sweetser Nov. 6, 1956 2,771,401 Shepherd Nov. 20, 1956 2,889,264 Spurlock June 2, 1959 2,901,417 Cook et al Aug. 25, 1959 2,917,448 Beuther et al Dec. 15, 1959 2,937,134 Bowles May 17, 1960 3,071,542 Davis et a1. Jan. 1, 1963