US1825976A - Process of cracking mineral oil - Google Patents

Description

Oct. 6, 1931. A. E. PEW, JR., E'r AL 1,825,976

PROCESS OF CRACKING MINERAL OIL l Filed NOV. 9, 1927 d "im I )Patented Oct. 6, 1931 ,Ni'rED STATES PATENT OFFICE ARTHUR E. PEW, JR., 0F BRYN MAWR, AND HENRY THOMAS, 0F RIDLEY PARK, .PENN- SYLVANIA, ASSIGNOIBS T0 SUN OIL COMPANY, 0F PHILADELPHIA, PENNSYLVANIA,

A CORPORATION OF NEW JERSEY IROCESS OF CRCKING MINERAL OIL Application led November 9, 1927. Serial No. 232,024.

In the artof treating high boiling and high gravity hydrocarbon oils for the purpose of decomposing them to convert them into lower boiling and lower gravity hydroF carbon oils, which art is generally known as' cracking, the most common procedure is to subject the oilto be cracked to high temperature and a substantial pressure above atmospheric. The art is a very Old one. In recent years, its most useful industrial apphcation is in the production of gasoline from such higher boiling and higher gravity petroleuin oils as gas oil, fuel oils, and kerosene distillates. The cracking operation may be conducted in batch or it may be carried on in a continuous manner. Q

In continuous processes for cracking higher boiling petroleum constituents to gasoline, it is old and well known to convey the oil to be decomposed, which may be initially preheated, through a long tube or tubes in immediate contact with heating gases and therein to raise the oil to a cracking temperature and either complete the cracking therein or perform a part of the cracking therein. In the latter case, the partly cracked oil may be conveyed to a container or reaction chamber of relatively large crosssection, wherein the pressure is maintained, and wherein the temperature of the oil falls, although not below an eective cracking temperature. Tn this chamber the crackingv is completed and the oil will be relatively quiescent to allow rdeposition of the coke which is formed under the high temperature conditions. The oil may be allowed to escape through a loaded valve into a vaporizer, wherein there is maintained a pressure but little above atmospheric. After passing the valve, most of the oil immediately vaporizes. The vapors are fractionated and condensed. Tf all the cracking be effected in the tubes, the formation and accumulation of coke in the tubes tend to insulate the oil from the heating gases and clog the tubes, necessitating their frequent cleaning. This process is therefore generally regarded as in ipracticable. If only part of the oil 1S cracked in the tubes and if the partly cracked oil is conveyed from the tubes to a large reaction chamber, in which the cracking is completed, as explained above, the trouble is merely ameliorated, in that some of the coke that would otherwise form or deposit in the tubes forms or deposits in the large reaction chamber. Frequent cleaning of the reaction chamber is therefore necessary. This is entirely practicable, but the drop in temperature that occurs in the reaction chamber slows down the reaction to a very great degree; it being possible to appreciate the extent of such slowing down if it be borne in mind that with every increase of temperature of about' 18 degrees F., the speed of reaction doubles.

The ideal cracking process is one wherein the o il is subjected to the cracking operation and is completely cracked while traveling along a long thin stream, from which it con-` tinuously escapes into an expansion chamber, the gasolineA being separated, fractionated, condensed and purified by known methods. The object of this invention is to provide a practicable process of this character. u Before describing the process, it may be informative to set forth more in detail the cause, nature and effects of coke format-ion.

With the usual method of heating the oil to raise it to a cracking temperature, whether it is flowing in a long tube or comparatively quiescent in a tank, the temperature d ierence between the furnace gases and the oil is and must be very great. Consequently the film of oil immediately in contact with the metal of the tube is raised to a temperature much above that of the average teinperature of the whole body of oil. This causes the highly heated part'of the oil to be raised not only to a decomposition or cracking temperature but to a temperature high enough to separate out certain constituents and convert thein into coke, which is deposited on the tube wall. Substantial deposition of coke does not occur until after the process has been in operation for some little time. Once, however, the coke starts to form, it forms very rapidly. The coke acts as an insulator, requiring more heat to penetrate it in order to heat the entire body of oil to a cracking temperature. This raises lrs the temperature of the tubes. The coke that has already formed acts to absorb oil, which, at the increased temperature of the tubes, causes it to form coke ra idly. In short, once the formation of the co e starts, its rate of production is constantly accelerated. Fur-- ther, the insulating effect of the accumulating layer of carbon in thetubes necessitates a still greater temperature difference between the furnace gases and the tubes and the subjection of the peripheral zone of the stream of'oil to still higher temperatures i n order that the main body of the travelingoil shall be subjected to the lower crac ing temperature desired.

It is clear that if the temperaure difference between the heating medium and the oil could be economically reduced, the condltions promotive of the formation of coke would be, to a substantial degree, eliminated.

We have found it to be practicable to secure the desired low temperature difference between the heating medium and the oil by '5 means of the indirect application of the primary heating agent and the direct transmissions of heat to the oil by means of a secondary heating agent which, under practicable absolute pressures, will boil at a temperature to which it is desirable to raise the oil, and which, in its vapor f orm, is f iowe into heat exchange relation with the oil and is condensed thereby and gives up its latent heat to the oil, the condensate returning to a liquid body of the substance, which is continuously being heated by the primary heating agent and is continuously generating vaor. In the application of Pew and rI `homas, o. 13,040, filed March 5, 1925,'wh'1ch is a continuation in part of an application filed March 15, 1924, No. 699,615, the advantage of using a vaporized metal, suchas mercury, as a direct heating agent for mineral oil, is clearly and fully disclosed. Mercury has a high boiling point and high heat conductivity; it will not oxidize or disintegrate when heated or brought into contact with iron and it may be condensed and its latent heat transferred to the oil by heat exchange at temperatures within the desirable zone of oil cracking temperatures. The most important and valuable quality of mercury vapor, in its application to the cracking of oil, is that, to accomplish the same degree and rapidity of cracking, the difference between the temperature of the mercury vapor and that to which it is necessary to raisel the oil need be very small relatively to the necessary temperature difference between the oil and ordinary furnace gases or other ordinary heating media. In other words, the temperature of the heating medium from which heat is directly transferred through the tube wall to the oil may be so relatively low that all the objectionable coke forming conditions liereinbefore recited are minimized.

It should be understood, however, that we do not profess to be the first to suggest the use of mercury vapor as a heating medium for cracking oil, our process also involving other features, in addition to the known process of cracking oil while it is flowing through a long tube, which features will be fully disclosed and explained in the following description.

In heating oil by means of any ordinary heating medium, such as furnace gases and very hot oil, a great temperature difference between the furnace gases and the oil is not objectionable while the oil is being heated up to a cracking temperature. Below about (say) 600 F., heat can be applied to the oil at a comparativel high rate per square foot of heating surfhce Without injurious local cracking. In our process, we preheat the oil, first, by heat exchange with the highly heated cracked oil, and second, by the heat of the furnace gases" after they have passed beyond the mercury boiler. Thereby the oil is preheated to a high temperature, which, however, preferably does not reach,

and ought not to much exceed, the vtempera-y ture at which substantial cracking begins. Either one of the described methods of preheating may be omitted, but if either be omitted, it is preferred to preheat by means of heat exchange with the cracked oil.

The oil, thus raised to near, or about, or slightly above, a low cracking temperature, is caused to circulate with great rapidity through a tube that is in heat exchange relation with mercury vapor, which is thereby condensed and gives up its laten-t heat to the oil. The temperature of the mercury vapor is sufficiently above the maximum temperature to which it is desired to heat the oil to effec-t an elevation in the oil temperature to about its maximum substantially before the oil has completed its passage through the tube and preferably after it has passed through only a minor part of the complete length of tube. Thereafter, the oil is held at about this maximum temperature by continuing in heat exchange relation with mer cury vapor at a reduced temperature not much above that to which the oil is raised in this3 first and minor part of the length of the tu e.

The process therefore involves a rapid but gradual heating of oil by heat exchange with mercury vapor at a relatively high temperature to the degree required to condense it and abstract its latent heat, and a subsequent maintenance of the oil at a high temperature, preferably for a longer period of time, by heat exchange with mercury vapor at a relatively low temperature approximating that to which the oil is initially heated by the mercury vapor.

ieaere To secure the conditions above specified, a subordinate feature of the process consists in providing one mercury chamber through which the first part of the oil stream flows, and one or more additional mercury chambers through which the remaining part of the oil stream iows, and in maintaining, in the first chamber, a relatively higher pressure (involving a higher temperature of condensation) and in the remaining chamber or chambers a relatively lower pressure (involving a lower temperature of condensation).

Other advantageous features of the process will be brought out in the following description.

While the-process is not dependent for its execution on any particular construction and arrangement of apparatus, the drawings represent the lay-out of a plant in which the process is adapted to be carried out in a particularly practicable and advantageous wa hyig. 1 is an elevation, largely diagrammatic, of a complete cracking plant.

Fig. 2 isa detail View of one of the heat exchange or cracking units.

The mercury boiler a, is contained in a furnace b which is provided with a down-take c. The furnace gases pass upward around the boiler, thence laterally and thence downward through the down-take c and thence up'- ward through a stack d.

Above the boiler is arranged a series of cracking units, each comprising a shell e, functioning as an oil heating and mercury vapor-condensing chamber, and a nest of tubes f. The tubes are so arranged that the oil entering from pipe g must flow back and forth repeatedly before leaving at i. It is desirable that the oil should be given suicient velocity toaid in cleaning the tube walls and the length of travel must be sufficient to allow proper time for the reaction to take place.

From the mercury boiler a a mercury vapor line z' extends upward and thence laterally along the cracking units 1, 2, 3 and 4, and is provided with four feeds 11, 12, 13 and 14 extending to the respective chambers e of the four cracking units 1, 2,3 and 4. Each feed is rovided with a Valve j.

rom the several chambers e extend mercury condensate return lines 21, 22,23 and 24 to a header p. Each return line comprises a trap consisting of a cup k and a goose-neck m, and another goose-neck n.. The upper part of goose-neck m connects with the upper part of the cup by a vapor pipe 0. The part n connects with a header p, which is connected with the mercury boiler by means of a pipe 1'.

In the down-take c is a nest of tubes s comprising a preheater. Oil is pumped into one of the bottom tubes of the preheater and thence flows back and forth repeatedly and finally leaves one of the top tubes and flows through pipe g to cracking unit 1. r1`he oil then ows successively through units 2, 3 and 4.

The oil to be cracked flows from the source of supply through line 30 (in which is interposed a heat exchanger 3l) to preheater s.

From the last cracking unit 4 the oil flows through a pipe a, (in which is interposed heat exchanger 31) into a vaporizer o. The exit to the vaporizer may be controlled by a valve u. Beyond the vaporizer may be avbubble tower fw and beyond that a condenser m.

A longitudinal section of one of the cracking units is shown in figure 2. The shell c must be made of sufficient strength to carry the required pressure of mercury vapor. 'lhe tubes f must be made of a size and strength suitable for carrying oil at the high pressure specified. For the liquid phase system, the tubes should have walls about halfV an inch thick. rlhe tubes may be welded to the heads of the shell e. By reason of the temperature difference between the tubes and the shell, provision should be made for expansion and contraction. One way of providing for expansion and contraction is to make the shell in two sections with'an expansion joint e at the center. These details, however, may be worked out by engineers skilled in the art of heat exchanging.

A more specific description of the operation follows: lt should be understood that the more detailed information, such as pressures and temperatures, are illustrative only.

Oil to be cracked is pumped through line 30 and in flowing through heat exchanger 3l is heated to a fairly high temperature by heat exchange with the cracked oil flowing through line la.. Thence it flows through the preheater s and line g and thence in series through the tubes f of the cracking units 1, 2, 3 and 4. It is permissible to omit, or cut out of the circuit, the preheater s and flow the oil direct from line 30, through line 32, to line g and thence to the cracking units. The oil leaving the tubes f of unit 1 through pipe L enters unit 2 and thence flows through units 2, 3 and 4 in series.

Mercury` vaporized in boiler a flows through line d and thence, through the independent feeds 11, 12, 13 and 14, in parallel to the mercury chambers e of the four cracking units. The temperature and pressure within any one of the mercury chambers e of the successive units may be accurately controlled by more or less throttling the valve on the independent feed. In each chamber e, mercury vapor is condensed and its latent heat transferred to the oil. Mercury condensate returns from chambers e through the return lines 21, 22, 23 and 24 to the header p and thence, through pipes r, to boiler a.

From the last cracking unit 4, the oil Hows through line l, and heat exchanger 31 (exchanging heat with the oil owing through line 30) and thence escapes into a vaporizer fv, wherein most of the oil is vaporized, the small remainder being thrown down. The vapors thence pass into bubble tower fw, wherein a part of the vapors is condensed, the vapors bubbling up through descending condensed oil. The tower may be constructed in accordance with the patent application filed by Iew and Thomas February 23, 1924, Serial No. 694,470, Patent No. 1,723,368. The vapors, comprising mainly gasoline, pass either direct to a condenser or may be subjected to a purification process on the way thereto, or subsequently to condensation.

In the heat exchanger 31, or in the heat exchanger 31 and prelieater s, the oil may be heated to any desired temperature, but should preferably be heated up to or above a low cracking temperature. In the first cracking unit the oil is raised to, or close to, the maximuni cracking temperature, which is preferably rom 830 to 900o F. The temperature of the oil is maintained at or about this tein- )erature in the succeeding cracking units.

n the case of atypical gas oil, the preferred crackin A temperature will be in the neighborhood o 860 F. This temperature is subject to considerable variation dependent on the length of the coils, the speed of travel of the oil therethrough and the character of the charging stock.

The boiler pressure may be maintained at about 142 pounds per square inch absolute pressure, under which mercury vaporizes at about 960 F. The temperature and pressure s cified are, of course, merely illustrative.

e can be varied, within reasonable limits, or a given stock and the most desirable temperatures and pressures will vary with different stocks.

Cracking unit 1 may be operated at the full mercury boiler pressure, whereby there will be established a temperature difference between the mercury and the oil as the latter leaves cracking unit 1 of from 45o to 70O F. In order to avoidV a progressive increase of temperature in succeeding cracking units, the valves in lines 12, 13 and 14 are more or less throttled, thereby reducing the pressure and temperature in the corresponding units 2, 3 and 4. It is also desirable to avoid a substantial diop in temperature in units 2, 3 and 4. The valves in lines 12, 13 and 14 can be so throttled that a substantially uniform oil temperature, corresponding approximately to that of the oil as it leaves vaporizer 1, will be maintained in units 2, il and 4. This will involve maintaining in the successive units a pressure of from about 5)() to 105 pounds absolute, giving a temperature of about 830o to 9000 F. In other words, the temperature of the heatng medium will be maintained at about the temperature of the oil. Actually, in order to avoid a drop iii temperature in units 2, 3 and 4, the temperature of condensation of the mercury vapor should be slightly above that of the oil so as to maintain a very small temperature difference between the inercury and the oil. The process, however, does not exclude the maintenance of the conditions that will efl'ect a slight raise, or permit a slight fall, in the temperature of the oil dur' ing its passage through unit 2, 3 and 4; the process contemplating merely an approximation to the maintenance of a uniform tem perature of the oil.

It should also be understood that valves on pipes 11, 12, 13 and 14 are provided merely for convenience. If the apparatus be carefully designed, as for example, by an ap propriate variation in the diameters ofthe pipes so as to give a reduced rate of flow of mercury vapor through pipes 12, 13 and 14 into the units 2, 3 and 4, valves can be dis'- pensed with; but their retention is highly desirable, in that they provide a means for controlling the pressure and temperature in the several mercury chambers, in case different charginf stocks ai'e run that require heating into different temperature ranges.

Each of the cups k, which communicates with the mercury condensate discharge pipe 21, 22, 23 or 24, connects with a goose-neck pipe m and forms therewith a liquid mercury seal or trap, in which collects any sediment that may form. Pipe 21 (or 22, 23 or 24) and the vapor space at the top of the cup 1.7, is connected, by means of pipe 0, with the inflow section of a pipe n, which section extends downward for a considerable distance and thence upward, for a distance equal to half the length of its downward extension, to the header p. The desirable depth of the U-shaped section of pipe n below the header p and the desirable length of the inflow section of pipe n above header p will depend on the pressure which it is intended to maintain in the boiler.

If it be assumed that the pressure in the mercury vapor chamber of unit 1 (or 2, 3 or 4) be the same as the boiler pressure, the two columns of liquid iii the U-shaped pipe section a will have a common level. If a reduced pressure is established in the mercury vapor chamber by throttling the inflow valve as hereinbet'ore explained), the liquid mercury will be forced down the right-hand limb of section a and up the left hand limb thereof until an equilibrium of pressures is established. As mercury condensate runs into the inflow end of pipe n, the levels of both columns will raise in unison until the righthand limb of section n is filled, after which the condensate will continually overflow into header p.

The cracking units, the header p, and the connections between them should be located at such height above the boiler that the column of mercury in the return line may have a head equal to about the diil'erence between vie Lezama the pressure carried in the boiler and the pressure in the last vaporizer (4) of the series.

A by-pass for excess mercury vapor is provided between the mercury vapor line i andthe liquid mercury return line r. In-

this by-pass is located a safety valve, or maximum pressure relief valve y, and a condenser a.

rEhe operation of the described elements 21, k, m, 0 and n, and g/ and z, is more fully set forth in applications filed by us May 29, 1926, Serial No. 112,443, Patent No. 1,714,811 and Serial No. 112,444, Patent No. 1,714,812.

'lihe critical temperature of a typical gas oil that is subjected to a cracking process is about 860 F. Such gas oil, when substantially filling a confined space, will be converted into a true gas at about such temperatune. Whether, in the present process, due to fractional vaporization of the oil as it is heated up toward the maximum cracking temperature, a complete conversion to gas occurs when the critical `temperature is reached, cannot be asserted with certainty, but it cannot be asserted that it is not so converted or that the cracking process is a liquid phase process. lf there is such complete conversion to gas, most of the cracking must occur in gas phase. Our process. however, is not limited to gas phase cracking, or vapor phase cracking, or mixed phase cracking, but covers cracking, of whatever nature it be, at a temperature distinctly above the minimum cracking range, and, in the case of a typical gas oil, preferably within a range of 830 to 900 F. and still more preferably within a ran-ge of 850 to 870 F. The preferred temperature will vary, also, with the character of the charging stock and the time which the oil occupies in traveling through the. coils.

In whatever phase the oil is in as it leaves the last vaporizer unit. it will be largely a liquid after it passes through the heat exchanger 31.

The pressure maintained on the oil in the system may be anywhere from 600 to 1500 pounds, or even higher.

ln the cracking process described, due to the small temperature difference between the mercury vapor and the oil. there is substantially no local heating of the oil above the highest predetermined cracking temperature, and therefore one condition promotive of the formation of coke is avoided. By providing a. suiciently great total length of tube, the oil can be made to iiow therethrough at a high rate of speed and therefore tends to scour from the inner wall of the tube any coke that forms thereon. By avoiding a substantial drop in temperature, as occurs in a reaction chamber, to which no heat is applied, located beyond the heating coil, the speed of reaction is not merely increased, but multiplied; it being well understood that the speed of a reaction, such as the cracking of ticularly adapted to the purpose of the present process and which may be considered, in the present process, as equivalents of mercury. It is also possible to achieve, in part, the purposes of the invention, by using any known heating medium, such as hot furnace gases if their temperature be so regulated as to maintain, in the later unit or units of the series, the maximum cracking temperature; because the maintenance of the maximum cracking temperature, at which the crack-i ing reaction is very rapid, throughout a long length of tubing, permits a speed of travel of the oil great enough to carry out with it a large proportion of the carbon as it is formed.

`With a heating medium such as furnace gas,

however, the maintenance of a relatively low temperature of the heating mediumis not possible and the process is therefore not workable to accomplish all the objects of the invention or to accomplish such objects with maximum eiciency.

Having now fully described our invention, what we ,claim and desire to protect by lLetters Patent is:

1. The process of cracking mineral oil which comprises establishing a owing stream of oil of great length and small thickness in series through a number of confined spaces, flowing mercury vapor into one or more confined spaces of the series in heat exchange relation with the iiowing elongated streamof oil, maintaining in said confined space or spaces such relatively high superatmospheric pressure of mercury vapor as to establish a temperature of condensation sufficient to heat the oil through a range of cracking temperatures to a temperature substantially in excess of 830 F. with resultant fractional conversion of oil to vapor and substantial cracking of the oil and with resultant condensation of mercury vapor, and flowing mercury vapor into one or more confined spaces in heat exchange relation with a more advanced part of the length of the flowing stream of oil and maintaining in such last named confined space or spaces a relatively low absolute pressure of mercury vapor but a pressure suiiiciently high to maintain the oil at a cracking temperature substantially in excess of 830 F. throughout substantially the greater part of the length of the oil stream.

2. The process of cracking mineral oil which comprises establishing a ilowing stream of oil of great length and small thickness in series through a number of confined spaces, flowing mercury vapor into one or more confined spaces of the series in heat eX- change relation with the flowing elongated stream of oil, maintaining in said confined space or spaces such relatively high superatmospheric pressure of mercury vapor as to establish a temperature of condensation sufficient to heat the oil through a range of cracking temperatures to a temperature in eX- cess of 830 F. with resultant substantial cracking of the oil and condensation of mercury vapor, and flowing mercury vapor into one or more confined spaces in heat exchange relation with a more advanced and substantially greater part of the length of the flowing streamof oil and maintaining in such last named confined space or spaces a relatively 10W absolute pressure of mercury vapor but a pressure suiciently high to maintain the oilat a cracking temperature in excess of 830o F. throughout substantially the greater part of the length of the oil stream.

In testimony of which invention, we have hereunto set our hands, at Marcus Hook, Pennsylvania, on this 7th day of November,

ARTHUR E. PEW, JR., HENRY THOMAS.

Images