US2420030A - Motor fuels and preparation thereof - Google Patents

Description

Mayfi, 1947. v R. c. BRANDON 2,420,030

MOTOR FUELS AND PREPARATION THEREOF Filed April 13, 194:5 2 Sheets-Sheet 1 FLOW SHEET FETRQLEUM LIQUID VA'POE PHASE CRACKING FRACTEC'NAL. DISHLLATION Lam-1T FRACTIONS 30 -l +5F EAVY FRACTIONS BELOW 89F. (41190-430?) .Anovz m5n HYmaoesNA-r ION FRACTIONAL. lS'l'lL-LATION .BELQW 90F 9o=-uor ABOVE |30F asoPENTAwE N'PE.HTANE :slzAuci-wrn HEXANES IOO :30"?

May 6, 1947. R. c. BRANDON MOTOR FUELS AND PREPARATION THEREOF Filed April 15, 1943 2 Sheets-Sheet 2 HYDROGEN ATE], Pnobuafs amma FRACTlON 90-l'+5F E N N A e T e m u n u 2 F w a a A a u" v N v T .u an H A N A .5. r r m a o: a a P T W H E o M Y... w. M m p .QN 3 l A; .2 i T R A H F C Au 0 AU 3 m a N m P I a I I l l II I I G R z m m o E E E IE5 E a E m. E m 5. NM 1 N B N E T N T "W: m. (m A a N T u a: N N 6 T x i2 3 a z v u B h 9 X ms nu H N 0 a 2.. s o 0 HP 2 L YW L B-H H FL 5 m f I m we 5, 2w 2 L T v c YY2 H .T Y E HH- T Y:- H m a T3 u m U. D Y HEW m H3 M C MME M x 3 M 3 2 3 2 n 2 B5 .7M, W'

Patented May 8, 194-7 MOTOR FUELS AND PREPARATION THEREOF Richard 0. Brandon, Elizabeth, N. 3., assignor to Standard Oil- Development Company, a corporation of Delaware Application April 13, 1943, Serial No. 482,832

. 2 Claims. 1

This invention relates to novel motor fuels and methods of preparing same and more particularly to a novel method of preparing an improved addition agent for motor fuels, especially for preparing cyclopentane in relatively pure form from petroleum raw materials.

It has been known for some time that cyclopentane is a suitable addition agent for motor fuels because it has a boiling point of 120.7 F., a Reid vapor pressure of about 9 lbs. per square inch, and an octane number (by the A. S. T. M. motor method) of about 85 to 90. Recently it has been found to be even much more valuable as an addition agent in leaded aviation motor fuels to be used in supercharged engines under rich mixtureconditions, because under such conditions it tremendously increases the power output of the engine, provided the fuel contains 3 cc. or more of tetraethyl lead (T. E. L.) per gallon. For instance, in the so-called 3-C blending value test for motor fuels, identified as Army-Navy test AN-VV-F-748 (for determining fuel characteristics in the supercharged engine with various fuel-air weight ratios), cyclopentane with 4 cc. '1'. E. L. per gallon shows a power output or I.- M. E. P. (indicated mean efiective pressure) of more than 450 lbs. per square inch (blending value in 20% by volume concentration) in an alkylate base stock, whereas the more desirable of the other saturated hydrocarbons boiling in the same vicinity and having a sufficiently high vapor pressure to be used as the volatility-imparting agent in a motor fuel blend, have substantially lower 3-C blending values, such as 290 for 2,3-dimethyl butane, 235 for neohexane (which is 2,2-dimethyl butane) and 218 for isopentane.

Thus it is one object of the present invention to provide a practical, economical and plentiful supply of cyclopentane for use as the sole volatility-improver in aviation motor fuel blends for supercharged engines. Such fuels require the use of special base stocks which per se have an A. S. T. M. octane number of atleast 85, and preferably at least 90, and a power output of at lease 180 I. M. E. P. in the 3-0 test, when containing 3 cc. of tetraethyl lead per gallon.

Suitable base stocks for the purpose of this invention are the hydrocarbon liquids made by alkylation of lower olefins (alkenes) having 2;

to 6, preferably 3 to 5, carbon atoms with isoparaflins (iso-alkanes) having, 4 to 5 carbon atoms, particularly isobutane having 4 carbon atoms. Such alkylates may be made by known methods such as by reaction in the presence of sulfuric acid or other catalyst, and generally have a boiling range of about -350 F. and preferably about '320 F., and usually havea-Reid vapor pressure of about 2to3. Another suitable base stock is the material known commercially as iso-octane but which is chemically referred to as 2.2,4-trimethyl pentane and which is made commerically by hydrogenation of di-isobutylene. This product has a boiling range in the vicinity of about 200-230 F. and 2. Reid vapor pressure. of about 2.

When preparing aviation fuels for supercharged engines, it is normally considered desirable to make a finished blend which has a Reid vapor pressure of about 5 to. 7 lbs/sq. in., because if the vapor pressure is much below 5, difficulties would be encountered in starting the engine, and if it is much above 7,.vaporlocking and other troubles may result. On the other hand, for maximum power output which is very desirable, if not essential, especially for military purposes, when an airplane takes oif, it is obviously desirable to incorporate into the motor fuelbase stock in addition to tetraethyl lead as much cyclopentane as possible (due to its high blending I. M. E. P. in alkylates of more than 450 lbs. per square inch) without exceeding the Reid vapor pressure limit of 7.

Now one of' the difficulties of the past has been that pure cyclopentane could be derived essentially only by the hydrogenation of a relatively pure cyclopentadiene obtained as a by-product during the destructive distillation of coal into coke, and the amount of it which can be obtained from this source is so insignificantly small that it is not only expensive but the amount available could not possibly supply the tremendous demand for motor fuel purposes. On the other hand, some cyclopentane has also been produced by straight distillation or fractionation of certain crude petroleum stocks in which it occurs by nature in relatively small amounts. Here again, however, there are two disadvantages, namely, that the amount actually available is still relatively small and furthermore it generally contains inseparably admixed therewith a small amount of a branched paraflinic hydrocarbon called neohexane (chemically referred to as 2,2-dimethyl butane, which has a boiling point of 1215" F., which is almost identical with that of cyclopentane itself, 1. e. 120.7 F.). It is believed that by even the best methods of fractionation which are available from a practical or commercial point of view, a cyclopentane product cannot be obtained which has a purity greater than about 90-95%. As shown in the following data, the small amount, i. e. -10% of saturated hydrocarbon impurities (perhaps neohexane) very greatly reduces the 3-0 blending value of such cyclopentane obtained by distillation from natural petroleum products and this reduction in 3-0 blending value may be very much greater in proportion than might be expected from the actual percentage of impurities present, because small amounts of impurities often exert a deleterious influence far greater than would be calculated arithmetically on a percentage basis.

Pure cyclopentane was subjected to a 3-0 blending value test in 20% concentration in an alkylate base stock having an octane number of 90, and with an addition of 3 cc. of tetraethyl lead per gallon. For comparison, the pure cyclopentane cut obtainable from fractionation of crude oil was similarly tested, except that 4 cc. per gallon of tetraethyl lead were used instead of 3 cc. The following table shows the test results:

drocarbons, fractionating the resultant cracked products to separate therefrom a fraction boiling Within the approximate limits of 90-145 F., hydrogenating said fraction under conditions to saturate the unsaturated hydrocarbons therein, and finally fractionating the hydrogenated product to separate therefrom a fraction boiling within the approximate limits of IOU-130 F. and consisting essentially of cyclopentane.

A primary feature of the invention is carrying out the cracking under conditions which tend to produce a relatively large amount of olefinic or di-olefinic hydrocarbons, the amount of carbon-to-carbon linkages to be split depending upon the molecular size of the raw material being treated, which may consist of various petroleum fractions such as naphthas, kerosene, gas oil, light and heavy lubricating oil stocks. If a very light stock is used as raw material such as a light naphtha containing a substantial amount of C5 hydrocarbons, the treating process may comprise essentially a dehydrogenation with little or no splitting of CC linkages. Some ring formation, i. e., conversion of chain compounds into corresponding ring compounds, is not objection able and in fact is desirable.

Although this vapor-phase cracking may be carried out by a batch operation and in various types of equipment known to the art, it is preferably carried out continuously by heating the hydrocarbon feed material, together with a diluent, as will be explained, in a pipe coil where 1 3-0 Blending Value Refracear Gyclopentane Bomnsggzange' tive Index E352 ASTM- Y o. N. cc. IMEP TEL/Gal.

Pure 120. 6 1. 4065 99+ 85. 0 3 425 Distilled from Crude 120. 2 to 121. 2 1. 4030 90+ 91.1 4 370 These tests show that the pure cyclopentane gave a power output (I. M. E. P.) of 425 with only 3 cc. of tetraethyl lead, whereas the narrow cut cyclopentane fraction distilledfrom crude petroleum only gave a power output of 3'70 even with 4 cc. of tetraethyl lead (the pure cyclopentane shows an I. M. E. P; greater than 450 when 4 cc. of tetraethyl lead are used). It is also interesting to note that these results are quite contrary to whatwould be expected from a comparison of the A. S. T. M. octane number of the plain cyclopentane, without any alkylate or tetraethyl lead, because the pure cyclopentane only had an octane number of 85.0, whereas the fraction distilled from crude oil had an octane number of 91.1.

Accordingly one of the primary objects of this invention is to provide a substantially pure cyclopentane having a purity substantially greater than 95%, and generally greater than 98%, and to provide this purer cyclopentane by a process which uses petroleum hydrocarbons as raw materials and converts them chemically into cyclopentane in large enough amounts for use on a commercial scale as a high power-output addition agent for supercharged aviation engines operating under rich mixture conditions.

Broadly the invention comprises, as outlined briefly in the flow sheet shown on the accompanying drawing marked Sheet 1, subjecting to vaporphase cracking a petroleum hydrocarbon stock composed of hydrocarbons having at least 5 carbon atoms, under conditions to produce substantial quantities of unsaturated S-carbon ring hyit is rapidly heated to a temperature of about 1200-1600 F., and maintained thereat for a relatively short time, such as less than 5 seconds, and generally less than 1 second, equivalent contact time at peak temperature. The diluent, calculated on a diluted hydrocarbon vapor basis, should be from 40 to mol percent of a suitable gaseous material such as air or steam. The use of a gaseous diluent of any type has the effect of reducing the partial pressure of the hydrocarbons present and therefore tends to increase cracking and unsaturation with relatively little tendency toward polymerization. The actual pressure at which the cracking is carried out should be in any case lower than about 200 lbs. per square inch absolute pressure, and preferably about 15 lbs. per sq. in. gauge, although, if desired, the operation may be carried out under vacuum, i. e., by the use of pressures ranging from atmospheric down to 1 lb. per square inch absolute pressure or even less,

After the desired amount of cracking has been effected, the cracked products are quickly cooled below the temperature at which harmful polymerization of the desired unsaturated materials might occur with any substantial speed, namely, to a temperature at least as low as 800 F. and perhaps down to 200 F., and this may be done by injecting water or unheated oil as a quenching medium into the transfer line as close as possible to the coil outlet, and/or fractionating the cracked products in a tower cooled by injection of water or 011.

After the cracked products have been cooled fraction being the one in which cyclopentadiene formed the cracking appear as well' as any cyolopentene similarly formed during the crack? mg This fractionation should be carried out as promptly as possible after the cracking so as to avoid unnecessary loss of cyclopentadiene by polym'eri'zation into dicyclopentadiene or higher polymers during storage,- which would, of course, boil at a substantially higher temperature thanthe 90445" F. fraction.

The next step according to this invention is the hydrogenation of the selectedcracked distilla'te fraction; This hydrogenation may be carried out in a number ofways,- but in any case should be carried out under relatively j mild hydrogenatirig" conditions with admixed hydrogen under a pressure of at least lbs. per square inch gauge so as to completely saturate the hydrocarbons srsefit without causing any splitting of C--C linkages, Ofie suitable. procedure is to hydrogenat'e' the 90-145 F. cracked distillate fraction with a Raney nickel catalyst or other suitable hydrogenation catalyst at a pressure of about 00'l,000lbs-. per square inch and ata tempera-- ture of about 200-500 F., preferably about 300- 3 5(l FL Other alternative hydrogenation procesures include the use of such catalysts as nickel sulfide-tungsten sulfide, molybdenum sulfide supported on a carrier such as alumina; clay, pumice, etc as-well as catalysts comprising oxides or sulfides ofthe group VI metals at temperatures of 50I0 800 R; and pressures of 50-500 lbsspersquare inchin' either batch or continuous equipment. As another alternative; the feed.- stock may be vaporized, mixed with; hydrogen, and passed oyer a catalyst in. a chamber heated to about 750 F.

The result of this hydrogenation step as will be seen more clearly by referring to: the accompanyin'g drawing marked Sheet 2, is to: efieot a spread inthe boiling points of thevarious hydrocarbons in such away as to facilitate the separation' of the primarily desired. cyclopentane from: all other hydrocarbons, except the small amount of neohexane which mayhavebeen presentinitially; or any that"woul'd be formed: upon hydrogenation-0f the correspondingolefinyrramely, the 3,3-dimethy1 butene-l. Howeven te'sts" do not show that any such olefin had been formed cracking, an etheoretically this'olefinh'as such a peculiar and unstable structure that it is improbable that any of it' wouldbe formed during the cracking. The accompanying chart (Sheet 2) shows clearly how hydrogenationseparates two" materials, one desired; and one undesired, which both boil at exactly the same temperature',; namely 10815" F., the piperylene being converted into pentan'e? and therefore changing in boiling point: from 108-5" F. down t6 96.8 F.', whereas the? cyclopentadiene changes into cyclopeiitanee with a rise in boiling point fronr1 0825." F. to. 1201 FL Thus hydrogenation has produced a sprea'zi' irr boilingrange of about 24 F.-betwee'n tha highly desired cyclop'entane andhighly uhdes'i-red normal-pentane, permitting ready Senna tionorthese two materials. It is also seen that ey'olopentene which boils at 114 which is 1ess thas 6 F2 from the plperylen'e} is alsd con verted' into cyclopentane and therefore is much more eas ly separated after hydrogenation, Al'- though effects just described are the most results of the Hydrogenation, it is also 6 noted: that several lowerboiling const tue ts; S ch as the: Zjmeth-yI-Zebutene whichboils-at l0l l E. and; hormone: which; boils at 93.4? l t, and therefore; are. both undesir bly ose to; normal pentone at, 96-.8:- F. and 2-pentene at- 97.391 1, are both: converted: by hydrogenation into isopentane which; boils at the muchlower temperature; of 82.4 Er, whereas the2-pentene is-converted i-nto pentane,- thus resulting in a spread of 14.4? F. betqween isopentane at $2.4 F.-,- which is a fairly desirable constituent for motor-fuelsg-andthenormal; pentanea-t-Q6-.8 R, which is very poor.- Also, all of; the, hydrocarbons boili-ng above cyclopene .tane-and neohexane in the'cracked; distillatefrace tiong are converted by hydrogenation into satu-rated hydrocarbons-boiling at aneven higher temperature; andtherefore are evenmore easily sep-' arated from cyclopentane after hydrogenation than before; These-heavier boiling hydrocarbons which boil between thelimiitsof l-2;8 Fl and 140 Fl before hydrogenation and boil betweenl-3 6 F. and 1.4;.6 F: aiter'hy r enation are allbranched hexanes whichhave fairly goodoctane number and fair power output blending valueand therefore areuseful asa mixed motor fuel blending agent;- even though they are notnearly as good-as cyclopen-tane' and therefore should desirably be separatedtherefrom;

As will be readily appreciatedfromthe chart, itmay be desirable in carrying out the present invention toselect a slightly narrower cracked fraction than one having; a boiling range of about 9Q-1l5 F.,-suchas'for instance -4 30 F.-, or preferably still about l00'-1'2 7 F., particularly if it is desired to obtain a relatively high proportionof cyclopentane without having to hydrogenate a large volume of hydrocarbons other than those leading to the formation of the; cy= c-lopentane or those which cannot otherwise be separated therefrom. In other words; by select ing a vapor-phase cracked fraction boiling between about- All) and 127 R, relatively little normal pentane will have to be handled, and only asmall proportion of isopentane resulting from the hydrogenation ofthe 2-methy1-2-butene,- but none of that resulting from isopr'ene; and little if any of th'e'branched hexen'es or correspond-ing branched hexanes boiling above: 128

Likewise in making the final distillation, 1; e2, after hydrogenation; it is preferred to cut the cy'clcpentane fraction at about IOU-130 F; or preferably at about LIN-425 F; was to recover substantially all of the cyclopntane in the hydrogenated product; without cbntaminating it with any of the normal pentane' or any of the branched hexanes boll-ingabove 130 F. By carrying' out the process of this invention a cyclopentane product has been produced which has a purity above98%' cyclopentane; andeven higher, and thisproduet hasbeen obtained in relatively high yields; for instance; ranging from 1 to 2% or higher based on the volume of original petro leuinlidiiids seen as aheavy' naphtha subjected to cracking.

In order to more fully explain the details as well as objects and advantages" of the" invention, the following experimental data are given? E sample- 1 A heavy virgin nannies having a tanner-sage of about 200 430" F.', whichwas obtained from F; and at a gauge pressure of about 15 lbs. per square inch, using 79.1 mol percent steam dilution and having an equivalent contact time at 1400 F. of about 0.86 second, the cracked products being immediately cooled by quenching with water injected into the fractionating tower in which the cracked products were subjected to distillation to recover a fraction boiling from the C range to about 145 F., final vapor temperature. This fraction was hydrogenated over a Raney nickel catalyst (made by leaching the aluminum out of a nickel-aluminum alloy with caustic soda) at '700 lbs. per square inch pressure and at a temperature of 315 F. The liquid product was analyzed for cyclopentane by analytical distillation and was found to contain about 25% by volume of cyclopentane. The fraction boiling at 120 F. was essentially 100% cyclopentane and amounted to 16% of the liquid product. The 113-129 F. fraction contained 95% cyclopentane and amounted to 24% of the liquid product. The cyclopentane boiling at 120 F. was not only easy to separate from the lighter fractions which include isopentane and normal pentane, as well as the heavier fraction which contained branched hexanes boiling above 130 F., but as shown by refractive index and other tests was practically absolutely pure cyclopentane. It showed no presence of neohexane which is the only saturated parafiln hydrocarbon which could be present in a fraction having a boiling point of 120 F.

In this experiment no attempt was made to recover the cyclopentane or other hydrocarbons which were lost in the gas during the cracking and hydrogenation, and assuming that the C5 portion of the lost gases contained the same proportion of pure cyclopentane as did the liquid C5 portion, it is estimated that the amount of cyclopentane lost in the gas, and which of course would be recovered in commercial operations, was substantially equal to the amount of cyclopentane recovered in the liquid hydrogenation products.

If these same yields are calculated on the basis of the amount of fresh heavy naphtha feed, originally subjected to cracking, it is estimated that the cracked fraction boiling from 90-145 F. amounted to 2.04% by volume, the C5 lost in the gas amounted to 1.43% by volume, the cyclopentane recovered from the liquid hydrogenated product amounted to 0.54%, the estimated potential cyclopentane lost in the gas is about 0.51%

:by volume. and the total estimated cyclopentane amounted to 1.05% by volume. This would mean that for every 100 barrels of heavy naphtha subjected to the cracking operations, 4 bbl. of material boiling at 90-145" F. could be obtained which on hydrogenation would produce 1 barrel of pure cyclopentane. This is a very great and unexpected improvement over the only other method used heretofore for obtaining cyclopentane commercially, namely, by distilling it from natural petroleum fractions, in which case very much more naphtha boiling at 90-145 F. would have to be distilled in order to obtain 1 barrel of cyclopentane fraction, and even the latter would not be as pure as the product obtained by the process of the present invention.

Example 2 The procedure of- Example 1 was repeated, except that the raw material feed to the cracking .unit was a light virgin naphtha and during the .cracking a 71.8 molpercent of air dilution was used with an equivalent contact time of 0.23.

Per cent by volume Recovered C5-145" F. V. T. cut 11.90 C5 lost in gas 5.86

Cyclopentane from hydrogenated Cs-145 F.

V. T. cut 0.73 Estimated cyclopentane lost in gas 0.98 Total cyclopentane 1.71

1V. T. means vapor temperature. This fraction contains about 69 wt. per cent C5; if only the C5 out had been hydrogenated, the cyclopentane from the hydrogenation would have been substantially higher.

Thus the total amount of cyclopentane recovered in Example 2 is slightly greater than that obtained in Example 1 when calculated on the basis of the total fresh feed to the cracking unit, since it was 1.71% in Example 2, compared to 1.05% in Example 1, but when based upon the amount of narrow cracked fraction subjected to hydrogenation (which was 11.90% based on the original feed) the yield of cyclopentane was only about 14-15%, compared to about 25% in Example 1.

The cracking conditions used in Example 1 had been selected as representative of conditions tending to give maximum aromatic production, whereas the cracking conditions used in Example 2 were representative of those tending to give maximum butadiene production. It will be obvious to those skilled in the art that considerable variation can be made in the cracking conditions to be used without departing from the general spirit and scope of the present invention, the primary requisite being reacting in the vapor phase at relatively elevated temperature with the use of about 40-90 mol percent dilution with air and/or steam or other suitable diluent, in order to produce a large amount of cyclopentadiene.

It is not intended that this invention be limited to the specific materials nor specific reaction conditions which have been given merely for the sake of illustration, but only by the appended claims.

I claim:

1. The method of preparing volatile liquid hydrocarbon products especially useful as a blending agent in leaded fuels for supercharged aviation engines to be used under rich mixture operation for maximum power output, which comprises subjecting to vapor phase cracking a petroleum hydrocarbon stock composed of hydrocarbons having at least 5 carbon atoms, under conditions to produce substantially large quantities of cyclopentadiene with other unsaturated C5 to C6 hydrocarbons boiling below 110 F. and above 127 F., fractionating the resultant cracked products to separate therefrom a fraction boiling within the approximate limits of -145 F., hydrogenating said fraction under conditions to saturate the unsaturated hydrocarbons therein, fractionating the resultant hydrogenated product to separate therefrom four fractions consisting essentially of the following four types of constituents, respectively: isopentane, normal pentane, cyclopentane, and branched hexanes, and

sired proportions to obtain a motor fuel blending agent having the desired combination of antiknock and vapor pressure properties.

2. The method of preparing a volatile liquid hydrocarbon product especially useful as a blending agent in leaded fuels for supercharged aviation engines to be used under rich mixture operation for maximum power output, which comprises subjecting to vapor phase cracking a petroleum hydrocarbon stock composed of hydrocarbons having at least 5 carbon atoms under conditions to produce substantially large quantities of cyclopentadiene with other unsaturated C5 to C6 hydrocarbons boiling below 110 F. and above 127 F., fractionating the resulting cracked products to separate therefrom a fraction boiling within the approximate limits of 90-145 F., hydrogenatig said fraction under conditions to saturate the unsaturated hydrocarbons therein, fractionatin the resultant hydrogenated product to separate therefrom a fraction consisting essentially of normal pentane, and combining the other components of the hydrogenated product essentially including isopentane, cyclopentane, and branched hexanes to obtain a motor fuel blending agent having the desired combination of antiknock and vapor pressure properties.

RICHARD C. BRANDON.

REFERENCES CITED The following references are of record in the file of this patent:

ruary 1944.

UNITED STATES PATENTS Number Name Date 2,279,721 Schmidt et a1 Apr, 14, 1942 2,324,073 Gaylor et al July 13, 1943 2,324,118 Sweeney July 13, 1943 2,052,148 Ocon Aug. 25, 1936 2,144,488 Forward Jan. 17, 1939 2,328,756 Thomas Sept. 7, 1943 2,330,069 Marshall Sept. 21, 1943 2,369,281 Chaney Feb, 13, 1945 2,371,794 Boyd Mar. 20, 1945 2,204,215 Greensfelder et al. June 11, 1940 2,360,555 Evans et a1, Oct. 17, 1944 FOREIGN PATENTS Number Country Date 463,244 British Mar. 24, 1937 542,500 British Jan. 13, 1942 OTHER REFERENCES Rec. Trav. Chim. 58, 1939, page 364. (Copy in Kudo et al., On the Hydrogenation of Cracked (Copy in 196-78.)

Aston et al., J. A. C. 8., vol. 65, pages 341-346, March, 1943. (Copy in Div. 31.)

Wilson et 2.1., Chemical Reviews, vol. 34, pages 1-50 (especially pages 3, 4, and 11), Feb- (Copy in Div. 31 at 260-666.)

Images