THE ISOLATION OF 2,3-DIMETHYL-8-ETHYLQUINOLINE FROM THE KEROSENE DISTILLATE OF CALIFORNIA PETROLEUM

Approved:

Approved: Dean of the Graduate School.

THE ISOLATION OF 2,3-DIMETHYL-8-ETHYLQUINOLINE FROM THE KEROSENE DISTILLATE OF CALIFORNIA PETROLEUM

THESIS

Presented to the Faculty of the Graduate School of The University of Texas in Partial Fulfillment of the Requirements

For the Degree of DOCTOR OF PHILOSOPHY

By Carroll L Key, 8.A., M.A, (San Marcos, Texas) Austin, Texas June, 1937

PREFACE

The experimental work described in this paper represents a part of an extensive study of the nitrogen bases obtained from petroleum distillates being carried out under the direction of Dr. J. B. Bailey.

This opportunity is taken to express to Dr. Bailey say sincere appreciation for his generous assistance and encouragement throughout the course of the investigation.

Carroll L Key

Austin, Texas April, 1937.

TABLE OF CONTENTS

Page

Introduction ........ 1

Experimental Part ...... 33

Conclusion 45

Bibliography ........ 46

INTRODUCTION I. History of the Search for Nitrogen in Petroleum

THE ISOLATION OF 2,3-DIMETHYL-8-ETHYLQUINOLINE FROM THE KEROSENE DISTILLATE OF CALIFORNIA PETROLEUM

The opportunity for beginning a fundamental and exhaustive investigation of the nitrogen bases present in petroleum distillates was made possible in the University of Texas Chemical Laboratory by a grant from the American Petroleum Institute. The Universal Oil Products Company and Mr. John D. Rockefeller jointly contributed 1500,000.00 for the promotion of purely scientific research in petroleum. Forty-one projects, to be completed in a five-year period beginning in 1936, were assigned to chemical and related fields of research. Project 20 of this program was assigned to the Texas Laboratory with Dr. J. R. Bailey as director under the title: Isolation and Investigation of Nitrogen Compounds Present in Petroleum. More than 116,000.00 was made available to the Texas Laboratory for Research Fellows before the exhaustion of this fund in 1931. The continuation of Project 20 would have been impossible without the interest shown and willingness to cooperate on the part of the Union Oil Company of California in processing and shipping a generous supply of any material requested..

A brief resume of the knov/ledge of the nitrogen basic content of crude petroleum and of petroleum distillates which had accumulated prior to the beginning of Project 20 seems pertinent. Necessarily,the late date at which man found use for petroleum and its products delayed any very exhaustive search for nitrogen, one of the minor constituents. The problem as to whether the bases were preformed or were the result of pyrolysis effected during distillation remained partially unsolved. The complexity of the basic content of petrolew distillates necessitated the development of effective methods of resolution before successful progress could be attained.

The earliest observation made of the presence of nitrogen in petroleum was that of de who noted that there was

a small amount of ammonia among the products obtained on exploding a mixture of the vapours of crude naphtha with an excess of oxygen in a eudiometer. Very little notice was taken of this discovery and no record is available of any research on petroleum for a period of fifty years. In 1868

observed the presence of nitrogen in bitvmins free from oxidizing agents,brought to the surface in the canons of Sulfur Mountain, California. His analysis on petroleum gave a nitrogen content of from 0.5645 per cent to 1.0855 per cent.

3 Bandrowski extracted Galician petroleum with 10 per cent

sulfuric acid and thereby demonstrated that nitrogen occurred in the basic form. He was unable to isolate any individual bases,but demonstrated that they would form insoluble platinic chloride salts, contained no oxygen and gave alkaloidal reactions .

The observation of the occurrence of substances in the petroleum of Saxony which gave alkaloidal reactions was mee by He called these substances pyridine-like because

of their odor, and similar results were obtained by acid extraction of oil from Boryslaw by Zaloziecki. 5 Caucasian pe-

troleum was investigated by Schestakow, who also found bases with a pyridine odor. It seems that the first systematic work toward the identification of these basic substances in petroleum was that made by on crude Russian petroleum. He

used the method of fractional crystallization of the platinic chloride double salts, but ,when bases were liberated, no single compound was identified and his calculations from pla~ tinic salts showed molecular weights of 104 to 309. Griffiths and Bluman obtained a basic oil boiling at 117° from Roumanian

petroleum; however, they were likewise unable to isolate any individual pure substance.

Progress of a definite nature in the determination of the pei' cent nitrogen in petroleum was slow due to the methods c used and to the interpretation of results obtained. When the

Dumas method w as employed, the results were generally high due to the presence of methane in the evolved nitrogen; however, it been found thit,by using two combustion tubes in sei’ies, results compare favorably with those obtained by the Kjeldahl method. 10 The results obtained by Mabery were duplicated in

11 the Texas Laboratory by Poth and others in a survey of North

and South American crudes. Table 1 shows the comparative results obtained from this investigation.

Obviously from these data it is noted that Californi petroleum furnishes the best source of supply for a study of the nitrogen basic constituents.

The earlier investigators believed that nitrogen in petroleum existed in the basic form. Evidently there was confusion,cue to lack of uniform treatment of samples and also as to the meaning of the word w crude.” That crude petroleum cortai. s bases which are readily removed by acid extraction the belief of Engler. Many other investigators believed pe-

troleum contains preformed bases which exist as such or in the form of their salts. isolated bases with a pyri-

dine-like odor; also found pyridine-like bases in the

solvent extract from Utah shale oil. He was inclined, to the belief that other bases occur,for he states: H None of the investigators lias accounted for the total nitrogen in various oils by the amount of pyridine bases extracted.” In 1932 Pyhala ls obtained bases (b.p. 75-200°) by extracting the kero-

sene fraction from Baku petroleum with sulfuric acid. He was inclined to the opinion that bases were not preformed, but arise during distillation by the decomposition of nonbasic nitrogen compounds. In the Texas Laboratory Poth and were not able to obtain any bases from California

crude by means of acid extraction.

Pabery l ? in his earlier investigation of petroleum bases

claimed that the nitrogen was in the form of tetrahydro Bz-1- 3, Py-S-trimethylquinoline, but in later work 18 contended that

the petroleum bases consisted mainly of quinolines and isoquinolines and were alkylated in positions 1, 2 and 3. Research in the Texas laboratory has proven this statement to be erroneous as to types of bases isolated and as to the positions which are alkylated.

Obviously then,it is fairly certain that prior to the investigation begun in the Texas Laboratory no definite claim can be made as to the isolation and identification of any particular base from petroleum.

1 de Saussure: Ann. chim. phys., 4, 314 (1817).

2 Peckham: Am. J. Sci., [3] 48, 250 (1894).

; Bandrowski, Franz: Monatsh., 8, 224 (1887).

4 Weller, A.: Ber., 20 , 2097 (1887).

$ Zaloziecki, Roman: Monatsh., 13, 498 (1892).

6 Schestakow, P. J.: Cheri. Ztg., 23, 41 (1899).

Chlopin, G. W.: Ber., 33, 3837 (1900).

c Griffiths, A. B. and Bluman, Bull. soc. ehim. 25, 725 (1901).

9 Day: Handbook of the Petroleum Industry, Vol. 1, p. 529

10 Mabery, C. F.: J. Soc. Chern. Ind., J 9, 505 (1900).

Poth, E.J., Armstrong, W.D., Cogburn. C.C., and Bailey, J.R.: Ind. Eng. Chern., 20, 83 T 1928;.

12 Engler: Chern. Ztg., 30, 715 (1906).

Weller, A.: loc. cit.

1 4 Day: loc. cit.

15 Pyhala: Chern. Ztg. , 46, 953 (1932).

both, et al.: loc. cit.

17 Mabery: loc. cit.

Mabery and Wesson: J. Am. Chern. soc., 43, 1036 (1920)

Petroleum Nitrogen Per cent California .102-. 815; av. .473 Heavy Mexican .358 Light Mexican .326 Mexican heavy mixed type .354 Colombian .226 Venezuelan mixed type Texas (Brazoria, Chambers, Galveston, Harris and Wharton counties) .231 .02-. 04 Texas (Jefferson, Liberty, and Orange counties) Kansas .040-.098 Louisiana .006—.090 Oklahoma .036-.165 Wyoming .017-.317 Pennsylvania .000-.000 Ohio .000-.000 West Virginia .000-.000

Table 1. Nitrogen Content of North and South American Crude Petroleum

II. Sources of Nitrogen Bases

letroleum Bases. — It lias been pointed out previously that the knowledge of the nitrogen content of petroleum and the condition in which this nitrogen exists was of a limited nature prior to 1926. The necessity for undertaking research problems in this field,such as created by the American retroleu Institute, was apparent. Project 20 was begun by making a petroleum survey (of. Table 1) which would, determine the crude contributing the largest nitrogen content from its distillates. California crude was selected for the investigation for it contains 0.473 per cent nitrogen. The problem of securing a supply of basic residues from the kerosene distillate of sufficient quantity for extensive research necessitates the cooperation of a large oil refinery. It was imperative, therefore, that the Texas Laboratory contact an oil company in California whose production was of such magnitude as would permit the collection of the required basic residues cheaply and in quantity. Several oil companies make use of liquid sulfur dioxide in processing the kerosene distillate. This operation is known as the Sdeleanu process and is primarily designed to remove the soot-forming olefines, aromatic hydrocarbons, sulfur and nitrogen compounds. The process consists in agitating the kerosene with one to one and one-half volumes of sulfur dioxide at lb-20 6 F. The separated layer, which is degassed at 160°F. and -10 lbs. gage, is the source of the basic material which was obtained for this investigation. The Union Oil Company of California uses the Edeleanu process at their Los Angeles Refinery, and kindly consented to furnish bases in quantity from the kerosene extract. A genuine interest in the whole problem of investigating the nitrogen bases of petroleum was manifested by Mr. R. S. Haylett, Manager of Production. He has made available to the Texas Laboratory a sufficient quantity of nitrogen basic residues to permit the carrying out of an extensive research program. The cost to the Union Oil Company has been conservatively estimated to be in excess-of

A Quantity of 50 liters of bases obtained from the sulfur dioxide extract of the kerosene distillate was shipped to the Texas Laboratory for the beginning of the research. These bases were furnished from the McKittrick crude containing 0.55> nitrogen. It was found that the total basic content represented 55> of the nitrogen present in the kerosene; consequently 45f was known to be bound in ine.rt form. After the completion of eight careful fractionations no one cut represented a pure compound, hence other methods of resolution had to be employed. The 275° zone yielded the largest volume of bases and from it through sulfuric acid treatment, 2,3,8-trimethylQuinoline, b.p. 278.9°, n 60/D 1.5828 was discovered. 3 * 9

These observers also found that a hot concentrated solution of picric acid in acetic acid would effect the separation of 2,3, 8-trimethylquinoline picrate from a hot acetic acid solution of the bases. On cooling the filtrate, a second picrate separated which proved to be that of the base having the empirical formula, b.p. 278.2°, n 60/D 1.4958. 20 These

two compounds represent the first pure nitrogen bases which have been isolated from petroleum.

The non-reactivity with benzoyl chloride, established the compound as a tertiary amine. Resistance to dehydrogenation excludes an olefinic linkage and hydroaromatic structure. Of importance are the several degradation products from nitric acid oxidation. The provisional structure assigned by Thompson and Bailey represents a so-called naphthenic base:

- t ; ight expected, the above structural interpretation proposed by Thompson has undergone drastic revision, in line with subsequent discoveries. found all attempts T-<J

to hydrogentite the base ended in failure as did attempts at degradation, but was able to make the phthalone and compound of the CigHggß base. The formula which Armendt pro* posed is;

In further study on reactions of the base, Lackey 22 was able to condense it with formaldehyde and through

oxidation and decarboxylation of this compound discover the new base, Proof of saturated behavior of the new base establishes the presence of two methyl groups in the original base. Lackey studied the action of cyanogen bromide and bromine on the base, but was unable to gain addi- tional information as to its structure.

One hundred grams of the base was sent to Dr. Homer Adkins, University of Wisconsin, for hydrogenation. He reported a consumption of 6 hydrogens in high pressure reduction at 300 atm. and 250 e and this was confirmed by analysis in the Texas Laboratory which proved the new product to have the composition Since it can be dehydrogenated to the original base, the conclusion follows that the molecule contains < pyridine nucleus. This was recently confirmed by a discovery by that pyrolysis converts the compound smoothly to

2,3,6-trinet hylpyridine.

On the assumption of naphthenic structure, the following formula accounts for the behavior of the kero base:

except 5 carbons in the form of undetermined alkyls must be as assigned to positions 5,6, or 7.

It is now certain the CpgHggN compound is of aromatic character, contains no bridging and has two reactive and one non-reactive methyl on the pyridine ring.

To place this investigation on a sufficiently broad basis, obviously more than 50 liters of bases would be required; hence the Union Oil Company agreed to furnish an additional three barrels of material from the kerosene distillate. It is seen that this meant a gigantic undertaking when it is observed that the kerosene fraction represents only of the crude petroleum, and of this is obtained as the Edeleanu extract. In order to obtain three barrels of bases from the kerosene fraction this operation then must represent the processing of 240,000 barrels of crude petroleum. The McKittrick crude from the San Joaquin Valley field was used.

This crude had a nitrogen content of 0.60. 0. R.

supervised the problem of securing the physical data on the three barrels of bases, and the data obtained are given in Table 2.

The rather exhaustive fractionation data were compiled and the bases were found to have a boiling point range from 210® to above 350® with about one-third of the material in pc the 270-290® range and the maximum amount at 285®/'’ It has

also been pointed out by Bailey and co-workers that the kero

423576

bases are of two types: aromatic 15-20 and non-aromatic 8G- e .

Since the isolation of the two bases, one of each type. there has followed the discovery of a number of others, largely 26 of the aromatic type, due to ease of separation. Biggs, '

working in the 265$ range using So g extraction method, isolated 2,4-dixn ethylquinoline, b.p. 274$ at 756 mm., and 2,3- dim ethyl quinoline, b.p. 273$ at 750 mm. In the 275° zone 2,4, s-trimethylquinoline, b.p. 280$ at 746 mm. was discovered by by means of his new method of separation, known us

cumulative extraction. This method, developed by Perrin as a means of separating the aromatics from the non-aromatics, will be discussed later. From the 253-256° fraction using the cu- PA mlative extraction method Lake isolated 2,8-dimethylQuino-

line, b.p. 252.4® at 754 w. A new naphthenic base of the 39 formula was encountered by while working on

the non-aromatic fraction in the 215-216° zone. This second non-aromatic base,b.p. 225.6° at 750 mm.,fails to form methiodide or phthalone, does not react with bromine and is unaffected when distilled with zinc dust. Armendt reported that it was completely saturated, but there is reason to believe that if reduction is attempted at 200 atm. and 250° with catalyst, as was done with base, hydrogenation may be accomplished’ The structural formula proposed by Armenct is:

If, as is indicated above, the base will not form a phthal one, and can be reduced, then it must be methylated on the F and the following structures my be more nearly correct than that suggested by Armendt:

Bratton, 30 in processing California cracked gasoline by means of amplified distillation isolated seven pyridine homologs, quinoline and quinaldine. Straight-run distillates in this temperature range yield a preponderance of non-aromatic bases, while Bratton encountered only bases of the aromatic type. These findings indicate that at high temperatures there results a dealkylation of polymethylated quinolines and a pyrolytic degradation of the polynuclear non-aromatics.

The only ethylated quinoline so far encountered in this research is 2,3-dimethyl-8-ethylquinoline, b.p. 284.8° at 755 mm., which was isolated by the writer from the 285° fraction (details to be given later). Results of further investigation on t e 2SO° range by Mr. Axe ci indicate that other ethyl

homologs may be encountered or the discovery of new tetramethylquinolines may be effected.

Another base of the non-aromatic type which is isomeric with Thompson’s K 2S N compound has been isolated recently by Mahan.Thompson’s base is optically inactive and work

is now in progress to determine whether Mahan’s isomer is optically active. This new base on pyrolysis yields the same pyridine as that obtained by pyrolysis of Thompson’s base.

The non-aromatics represent 85> of the total volume of bases isolated, and, as has been demonstrated, can be pyrolyzed to 5p pyridines and 80p quinolines. It now appears that these supposed naphthenic bases are largely hydroaromatic. Thompson’s and probably a few others are naphthenic in structure.

In processing the bases in the 200* range of the kerosene distillate by means of aoueous dulfur dioxide Meadows has

isolated 2,3-dimethylpyridine, 3,5-dimet hylpyridine, 2,3,5-trimethylpyridine, 3-methylpyridine, and two others which appear to be 2,4,6-trimethylpyridine and 3-methyl-8-ethylpyridine.

Bases from Bones.—Probably,the first source of nitrogen basic material investigated was that which was derived from the oil obtained in the destructive distillation of bones. Oippel made this study in 1711 and determined that bone oil had a high basic content. Bore than 100 years later Anderson

isolated from H Dippel’s oil® the first pyridine base ever obtained fro® a natural source. In 1880 Heidel and Ciamician, using the sulfuric acid extraction method, made the most complete study of the basic material found in bone oil. They processed the acid extract by boiling with potassium hydroxide which liberated ammonia, hydrolyzed the nitriles, and liberated volatile bases of pyridine and quinoline types. Table 3 indicates the numerous products isolated.

Coal Tar.—The interest in the study of the products obtained from coal has continued to increase since the work of in 1846. He reports finding considerable amounts

of pyrroles and pyridines from acid treatment of coal tar. It is evident that he worked at low temperatures and did not effect a very great pyrolysis of the materials. The many theories advanced as to the formation of coal all include means of accounting for the nitrogen content and nitrogen base origin. These theories sake room for bacterial action in the presence of moisture, the decomposition of plant and animal matter, and the action of chemicals with temperature and pressure effects, which may account for the small amount of hydrogen and oxygen found in coal and coal tar. The nitrogen content of coal has 37 been investigated by Fieldner, Selvig and and they re-

port 0.32 to 2.155 k As to the nature of the types of nitrogen compounds which result from coal tar studies,

states:

It has been suggested, with support of experimental evidence, that two forms of nitrogen compounds are present in coal, one in small amount of an amino nature easily decomposed by heat, the other ring formed, probably of the five membered pyrrol type, but highly complex and very stable until heated to a high temperature.

Because of the interest in the products obtained from destructive distillation of coal an extensive amount of research has been undertaken to determine the effect of temperature on the types of products finally isolated. Pyrolytic effects are unavoidable at high temperatures which result in

the formation of the simple pyridine and quinoline homologs from the more complex alkylated cyclic compounds. High temperature effects a large yield of benzene, inasmuch as toluene and xylene are dealkylate" , Since the major portion of basic material has been reported as simple methylated pyridines and quinolines, it is evident that high temperatures have removed long side chains and effected the decomposition of complex ring systems.

Morgan and Soule report that the density of the coal

tar obtained at high temperature distillations is great ex than that resulting from low temperature distillations. Their conclusions were incorrect, however, for they imagined hydrogena- Irion taking place in the low temperature range. Probably dehydrogenation at high temperatures produced the denser distillates. More recent data on temperature effects on coal tar 40 are those reported by Brown and Cooper from their studies

on light oil and tar resulting from heating Utah coal with superheated steam up to 725 c . They obtained hydrocarbons boiling from 2C c to 200 e which on fractionation resulted in the following: amylenes 30>, pentanes saturated and unsaturated of open and closed chain systems and olefinic and cyclic heptanes 17>. A very large yield of phenols was obtained while only a trace of pyridine resulted.

The pyridine bases are obtained in commercial practice today from the light oil fraction and consist of from 1 to Sy of this fraction. After the removal of the phenols the acid tar which has been steam distilled is treated with alkali to liberate the bases. There results about crude pyridine bases which are marketed in this form.

Shale Qi 1. —Extensive studies have been undertaken to determine the quantity and character of bases which may be obtained from shale oil. Oil-bearing shale may yield as high 41 as 50 pounds of bases per ton of shale. Low carbon shales

evolve the nitrogen as ammonia more easily than high carbon shales. The ammonia output can be increased, as in the case 42 of coal, by steaming out the residual coke. Most invest!-

gators report finding only pyridine and its homologs; however, Robinson and co-workers noted a series of compounds boiling from 270® to 390® and of the composition were unable to isolate any individual bases, but certainly must have encountered bases which would correspond to quinoline and its alkylated homologs, lower temperatures used in coking evidently account for the presence of the high boiling bases reported.

The most extensive work on shale bases was effected by Eguchiprocessed the oil from Fushun shale which

contained from 2to nitrogen bases. All bases isolate were of the tertiary amine type and 600 liters of tar which he processed, by acid extraction, yielded 3550 cc. of bases, b.p. 100° to 280®. The pure bases isolated from this material axe listed in Table 4. It may be of interest to give the formula which Eguchi deduced for calculating the boiling points of pyri dines:

T « 115.5 + 14 m + 28 m* + C

where T is calculated boiling point m is number of alpha methyl groups m* is number of beta or gamma methyl groups C is 4® for 3,3 substitution, 8® for 3,4 substitution.

The value of 0 is halved in the case of two side-chains in the 1,4 or p-position.

Cottonseed Meal.—ln recent work in the Texas Laboratory by barker, Gutzeit and Bratton^ 6 in an attempt to shed some light on the origin of petroleum, 337 pounds of bases obtained from the distillation of 23 tons of cottonseed meal with nitrogen-free lubricating oil as a liquid medium, were processes . This basic material was prepared by the Union Oil Company of California through controlled low temperature pyrolysis. If it be assumed that petroleum is of vegetable or animal origin, obviously carbohydrates, fats and proteins would be the starting mat erial in the formation of the bases. Whereas the bases so far encountered in petroleum are 80 to 900 non-aromutic, these compounds might be expected to form in the pyrolysis of cottonseed meal, if it is to contribute information leading to the genesis of petroleum bases.

A very complex mixture of basic material, of which only 50> was petroleum ether soluble, was recovered from the pyrolysis of cottonseed meal. It is of interest to note that the higher boiling fractions gave evidence of containing more than one nitrogen per molecule. These bases were vacuum distilled in an atmosphere of nitrogen; however, they were unstable and darkened on standing. Kight hundred seventy-six ml., b.p. 135 e to 172®, of petroleum ether soluble bases was available with n 25/b values between 1.4947 and 1.4992; while in the 172-215* range they obtained 2468 cc. of bases with n 25/$ values of 1.5010 to 1.5375.

The two fractions were resolved by means of amplified dis-47 48 tillation and cumulative extraction. These investigators

isolated pyridine and seven pyridine homologs from the low boiling fractions, and quinoline, isoquinoline, quinaldine, lepidine, 2,3,8-trimethylquinoline and a diazine, the higher boiling fractions. These compounds are shown in Table 4. The author*s main object may not have been attained; i.e., that of finding non-aromatics which are present in petroleum distillates, but the research gave ample evidence as to the origin of the aromatic bases. Probably then the complex compounds formed by bacteriological action at ordinary temperatures may result in nitrogen base formation when subjected to i . 49 pyrolysis.

In the search for the key leading to information as to the origin of the bases obtained from petroleum distillates, a study was made of the pyrolysis products of such materials as wool, eggs, casein, and silk fibroin. Attention is called to these experiments by Michelman, and Blioke and Powers. 0

By the hydrolysis of casein in the presence of formaldehyde, followed by distillation with lime, there resulted a yellow oil which contained primary, secondary and tertiary mines. From the mixture pyridine, 2,6-di met hyl pyridine,

isoquinoline, 4-methylisoquinoline, CyHgN, and were identified. Upon carrying out the experiment without the use of formaldehyde, no bases were formed. Bratton^ 3 repeated

this work to determine if quinoline could be formed; how ever, his results were negative.

kA In the pyrolysis of ovalbumin Pictet and Cramer isolated

acetic, propionic, butyric and succinic acids and a base, which they supposed was 1,2- dihydroaniline. The main product encountered was isocapronamide.

The pyrolysis of silk fibroin yielded phenol, p-cresol ;

Jr* F* indole, quinoline, aliphatic primary amines and pyrrole. 00

Compounds which have been isolated from various sources are given in Table 4.

References to abote Table

1. Spielmnn: Constituents of Coal Tar, pp. 163-192. 3, heap, Jones, and Speakman: J. Am. Chern. Soc., 43. 193 C (1921). 3. Ahrens and Gorkow: Bex., 37, 2062 (1904). 4. Ahrens: ibid., 29, 2996 (1896). 6. Eckert and Loria: Monatsh., 38, 225 (1917). c. de Coninck: Bull. soc. chim., 12] 41, 249 (1894). 7. Ahrens: Ber., 28, 745 (1895). 8. Ganguli and Guha: J. Indian Chern. Soc., 11, 197 (193.). 9. Eguchi: Bull. Chern. Soc. Japan, 2, 176 (1927 J. 10. McKee: Shale Oil, p. 116. 11. Staupper: Columbia University, 1926. 12. 'iggs a nd Baileys J. Am. shem. soc. , 55, 4141 (1933).

13. Leake and Bailey: ibid., 55, 4143 (1933). 14. Poth et al.: ibid., 52, 1239 (1930). 15. Ferrin and Bailey: ibid., 55, 4136 (1933). 16. Armendt and Bailey: ibid., 55, 4145 (1933). 17. Thompson and Bailey: ibid., 53, 1002 (1931). 18, Bratton and Bailey: ibid., 59, 175 (1937). 19. Gutzeit and Bailey: ibid., 58, 1101 (1936). 20. Parker and Bailey: ibid 58, 1102 (1936). 21. Oparina: Ber., MB, 566 (1931). 22. This work. 23. Eguchi; Bull. Chew. s oc. Japan, 3, 227-233 (1928). 24. Dahan: Unpublished work, University of Texas. 25. Weidel and Ciamician: Ber,, 13, 65 (1880). 26. Dinwiddie: Thesis, University of Texas, 1935. 27. leadows: Dissertation, University of Texas, 1937. 28. Pictet and Chou: Compt. rend., 162, 127 (1916). 29. Pictet and Cramer: Helv. Chim. Acta, jg, 188 (1919). 30. nohnson and Das chavs ky: J. Biol. Chern., 62, 197 (1924-5); J. Am. Chem. Soc., 41, 1147 (1919).

The proof of structure of the G x3 H the isolation of which is described ih the Experimental Part was established, first, through degredation experiments and, finally through synthesis. At first 2,3, 4-trimethylquinoline was suspected but this possibility was eliminated lay a mixed melting point of the picrate of the new base with the pi orate of an authentic sample and by analysis which established a composition agreeing

with rather than

Through chromic acid oxidation an acid of the formula iqNCCOH was foamed which on decarboxylation yielded 2,3-dim ethylquinoline, a kero base, which had been obtained by Biggs.This established the presence of an ethyl in the

original substance and its position at 8 followed from the fact that the above acid proved identical with the product obtained by King in the oxidation of 2,3,8-trimet hylquin-

oliae.

new base was obtained when 2,3*dimethyl-8-ethylquinoiine was condensed with formaldehyde, which through nitric acid oxidation, followed by decarboxylation of the resulting acid, yielced 3-m ethyl-8-ethylquinoline.

2,3-Dimethyl-8-ethylquinoline reacts normally with phthalic anhydride and like 2,3, 8-trimethylquinoline, which is of analogous structure, does not form a methiodide at 100°.

Attention is directed to the fact that none of the wellknown binuclear bases from coal tar, including quinoline, isoquinoline, quinaldine and lepidine has been encountered among the products occurring in straight run petroleum distillates; on the other hand, structures have been established for seven poly-alkylated kero quinolines. Such products up to this time have not been produced from a natural source other than petroleum. KQ However with two except ions the above quinoline homologs

were known to synthesis in advance of their isolation in the Texas Laboratory.

Another observation of scientific interest is, all kero quinolines, so far identified are methylated at position 2, the other positions of alkylation being 3,4 or 8. In all of these substances, positions 5, 6 and 7 are unsubstituted. it present a search in base fractions beyond the 285° boiling range for 2,3, 4-trimethylquinoline 60 and 2,3,4,8-tetram ethyl-

quinoline is in progress by W. N. Axe. In case these products are encountered, then the occurrence in straight run petroleum distillates of all theoretically possible methyl homologs of quinoline will have been established, where, in all cases, there is one methyl at position 2, with one up to three methyls at 3,4 or 8. Finally, it may be stated that 2,3-dimethyl-8- ethylquinoline, described in this thesis, is the only ethyl homolog of quinoline which has been obtained from a natural source other than petroleum. For the sake of clarity, the numbering employed is brought out in the subjoined structural formula:

” E :i and n Py H in this formula are the common abbreviations for benzene and pyridine, respectively. It now appears probable that none of the few remaining kero quinolines, to be isolated, will be found substituted in the Bz-nucleus at 5, 6 or 7; as a matter of fact, only a restricted boiling range, i.e., above 285° to 300 6 remains for a search of such products. Beyond this limit refractive indices of the aromatic constituents point clearly to substances of a higher polynuclear order than quinoline. Since acridines, because of their exceptionally high boiling points are excluded. The probability is, the kero bases concerned here are some of the following types:

In conclusion it nay be pointed out that petroleum offers a wealth of base intermediates for production of phthalone dyes which are of especial value because of fastness to light and chemicals. W. N. Axe has found that mixtures of the supposed trinuclear bases form phthalones which dye silk and wool a brilliant glistening brown, in contrast to the yellow color produced from quinolines methylated at position 2. Therefore, it is obvious, how research in the Texas Laboratory has unearthed a host of dye intermediates, destined to assume industrial importance in the near future.

19 Poth et al.: J. Am. Chern. Soc., .53, 1239 (1930).

20 Thompson and. Bailey: ibid., £3, 1002 (1931).

Armendt and Bailey: J. Am. Chem. Soc. , 55, 4145 (1933).

Lackey, R. W.: Dissertation, University of Texas, 1934.

23 ” * Pahan: Unpublished work, University of Texas.

Lake, G. R.: Thesis, University of Texas, 1933, p. 9.

25 Ibid.

Biggs and Bailey: J. Am, Chern. Soc. , 55, 4141 (1933).

27 Perrin and Bailey: ibid., 4136 (1933).

28 Lake and Bailey: ibid., sg, 4X45 (1933).

29 Armendt and Bailey: ibid., 55, 4155 (1933).

Bratton and Bailey: J. Am. Chen. Soc., 59, 175 (1937).

Axe: Unpublished work, University of Texas.

Mahan: Unpublished work, University of Texas.

Meadows: Unpublished work, University of Texas.

Anderson: Ann., 60, 86 (1846).

Wei del and Ciamician: Ber., 13, 65 (1880).

Anderson: loc. cit.

° Fieldner, Selvig and Paul: U. 3. Bureau of Mines Bull. No. 193, 1922.

$8 Porter: Coal Carbonization, p. 96.

Morgan and Soule: Chern, and Met. Eng., 26, 977 (1922)

Brown and Cooper: Ind, Eng. Oh era., 19, 26 (1927).

41 McKee: Shale Oil, p. 15.

42 Beilby: J. Soc. Chern. Ind., 6, 316 (1884).

Robinson: Trans. Roy. Soc. Edinburg, 28, 561 (1879).

44 Eguchi: Bull. Cheia. soc. Japan, 2, 176 (1927). 45 Eguchi: ibid., 3, 227, 233 (1928).

46 Parker, Gutzeit, Bratton and Bailey: J. Am. Chern. Soc., 58, 1097 (1936).

47 ~ Bratton, Felsing and Bailey: Ind. Chem., 28, 424 (1936), Ferrin and Bailey: J. Am. Chem. Soc., 55, 4136 (1933).

Day: op. cit. , p. 530.

50 Kichelman: Ind. Eng. Chern., 17, 471 (1925). 51 Blicke and Powers: ibid., 19, 1334 (1927).

Pictet and Chou: Compt. rend., 163, 137 (1916)

°° Bratten, A. 0. : Dissertation, University of Texas (1936), p. 18.

54r Pictet and Cramer: Helv. Ohim. Acta, 2, 188 (1919)-

Johnson and Daschavsky: J. Biol. Chern., 68, 197 (1924-5); J. Am. Chern. Soc. , 41, 1147 (1919).

* Number indicates literature reference.

66 von Braun: Ber., 57, 387 (1934).

Biggs and Bailey: J, Am* Chew, Soc., 55, 4142 (1933).

58 King and Bailey: ibid., 53, 1249 (1930).

55

von Braun, J.: Ber., 57, 387 (1924).

Thompson’s proposed structure

Structure for 'base

Bbl. 1 Bbl. 3 Bbl. 3 Boiling point 252 281 290 Sp. gr. .951 .971 .925 n 25/& 1.5208 1.5323 1.553: Sulfur .21f .30 Mt ro gen 7.70 7.70> 6.80 Hydrocarbon oil 10 4r . 3.0

Table 2. Analysis of Nitrogen Bases from Kerosene

Larger Yields Smaller Yields As nitriles: Methylamine Butyric acid Ethyl amine Valeric acid Aniline Caproic acid Pyridine Isocaproic acid Picoline Capric Lutidine Palmitic acid Quinoline Stearic acid Phenol Pyrrole Propionic acid as nitrile Methyl pyrrole Valeramide Dim ethyl pyrrole Toluene hydrocarbons: Ethylbenzene Naphthalene S10H16 C 11 H 18

Table 3. Products from Bone Oil

Base Coal Tar Shale Petro- leum Cotton seed meal Bones Proteins Aniline 1* 9 25 Dihet .Aniline 1 Pyrrole 1 10 ? 30 Incole 1 30 Carbazole 1 Ph enyl- p-naph- thyl carbazole 1 Pyridine 2 8 18 25 3-B-pyridine 2 9 18 18 25 3- -pyridine 2 23 27 19 4-- -pyridine 1 23 18 19 . * 2,3- B. r. py r i di n e 3 23 27 3, 4-D J .pyridine 2 9 18 19 2,5-D.1 .pyridine 3 23 18 3,6-IM .pyridine 21 9 18 18,19 28 3,4-D4.pyridine 4 23 3,5-DJ .pyridine 3 28 18,27 19 3-1t.pyridine 19 4-Et .pyridine 6 3 , 3, 4-1.11. pyridine 5 3,3,5-T.1. pyridine 21 23 27 2,3,6-T.1. pyridine 5 23 3,4 3 5-T.1.pyridine 4 23 3,4,3-T.l .pyridine 7 9 18,27 19 25 2-. -4-Et. pyridine 5 23 3-1 -6-Et .pyridine 3 9 3-1 -5-ht.pyridine 11 3-1 -5-Et.pyridine 27? 4-1 -v-Et.pyridine 5 2,3,4,5-Tet.H. ” 7 2,3,4,6-Tet.M. * 23 Base Coal Tar Shale Petro- < leum 3otton seed meal Bones Proteins . . —4— . — 'oyricine 2* 3,-D.K.-6-Et.- 23 pyridine Quinoline 1 23? 18,26 20 23 30 Isoquinol ine 1 20 28 4-Ket. " 28 2- .quinoline 1 18 20 4- .quinoline 2, 3-0. M. quinol ine 1 12 20 2,4—l.K.quinoline 13 2, -1.1 .quinoline 5,8-D.l .quinoline 8 13 2,3,8-^. M. quinoline 14,26 20 2,4,8-T.M. quinoline 2, 3-E.l'.-S-Et 15 quinoline Acridine 1 22 Hy d ro -a c r i di n e 1 16 016H 36 N 17 C16>W c 11 e 14 K 2 24 20 28 11 28 °12E 13 N 28 1 30

Table 4. Nitrogen Compounds Isolated to Date

EXPERIMENTAL PART

The 285* fraction of kero bases was selected for this investigation because preliminary research had revealed in this temperature zone a quinoline homolog in relatively large amount

Since all crude bases, as received from the Union Oil Company, contain considerable hydrocarbon oil, base-fractions . g between 283° and 287° (5.7 liters), which had undergone five ordinary distillations through an adiabatic column, were treated with a slight excess of dilute sulfuric acid. The supernatant layer of hydrocarbon oil was run off and. then, to insure removal of traces of non-basic admixtures, the solution was steam distilled and the bases were liberated with sodium hydroxide.

In the next step, the bases, in a slight excess of 1:1 hydrochloric acid, were processed through cumulative extrac* 61 tion* The aqueous layer containing the aromatic hydrochlo-

rides was washed five times with chloroform, in order to remove as far as possible non-aromtic hydrochlorides; in this way a high degree of segregation of the desired aromatic basetypes was attained. The non-aromtic residual mt erial is being reserved for future investigation. Table 5 applies the data on cumulative extraction.

In order to effect a further resolution of the bases after cumulative extraction, two additional distillations from a small still with more efficient regulatory control were carried out. The oil bath temperature ranged from 160® to 192® and the distillation progressed gradually from 100® to 130® at a pressure of sto 7 mm. The data presented in Table 6 reveal that an effective spread in boiling range and refractive indices resulted. Where distillations are necessarily carried out at low pressures to avoid deoomposition, fractionation is not so complete as where higher temperatures at atmospheric pressure are allowable. Therefore, it was desirable to further resolve selected fractions listed in Table 6; since it has been observed in the Texas Laboratory that even where mixtures of petroleum bases are exhaustively processec through fractional distillation, no cut ever represents solely an individual substance. However, each component base occurs in highest concentration in a distillation fraction boiling 3® to 5® lower than the base.

Since preliminary experiments had revealed 285° as the boiling point of 2,3-dimethyl-8-et hyl quinoline, fractions 12-2'5, inclusive, with boiling points of 280® to 286®, inclu-62 sive, were selected for amplified distillation."' Fraction

18, b.p. 283.5% is cited to exemplify the procedure followed.

Hydrocarbon oil (1000 ml.) boiling uniformly from 360® to 290® was admixed with 50 ml. of bases and the solution fractionated by use of the latest type still. In all, 31 cuts of 34 ml. each were collected. The bases, in each case, were extracted with a slight excess of 6 M sulfuric acid, all traces of hydrocarbon oil were steamed out, the solution was made alkaline with sodium hydroxide ,and then the bases were distilled over in steam. l&ch base-fraction was converted to picrates and a partial separation was effected by recrystallization from alcohol and glacial acetic acid. The bases were liberated in the usual way with sodium hydroxide. Distillation data are given in Table 7.

The above procedure gave very satisfactory results and was used on fractions 12 to 25 inclusive. From the original 5,7 liters of bases processed, 200 grams of pure 2,5-dimethyl-8-ethylquinoline was obtained. Had the tailings been reworked, this yield could have been increased materially.

Through recrystallization of picrates made from the numerous fractions, this salt of 2,3-dim ethyl-8-ethylquinoline, melting at 220°, was finally obtained in pure form. The min admixture at the start was 2,3,8-trlmet hyl quinol Ine picrate. The difficult solubility in alcohol of the acid sulfate of the base was employed in final purification. The fractions containing this substance in highest concentration were 22 to 28, inclusive, whereas fractions 2 to 12, inclusive, were rich in the base. From the higher boiling fractions a pure picrate melting at 196$ was separated, but not further investigated. These results demonstrate in a conclusive way the efficiency of amplified distillation.

The regained and distilled base, after being dried over solid potassium hydroxide, had the following physical constants: a.,;. 38.55; b.p. 284.6$ (755 mm.); d 20/4 1.017; n gQ/B 1,5940. From petroleum ether, the base crystallizes in large hexagonal plates. Characteristic of this quinoline is the non-formation of a quaternary salt with methyl iodide at waiter bath temperature.

Anal. Calcd. for C, 84.32; H, 8.10; N, 7.57; Found: C, 83.94; H, 7.89; M, 7.54; Mol. wt. subs. 15.50 mg.; Camphor 155.55 mg. dt, 21$. Calcd. for 185; Found: Mol. wt. 189.

He rate.—This salt, which is difficultly soluble in the common solventscrystallizes from glacial acetic acid in slender yellow needles melting with decoatposition at 220$ .

Anal. Cal cd. for C 19 H 18 0?N4: C, 55.07; H, 4.35; B, 13.53. Found: C, 55.19; H, 4.34; I, 13.46.

Fit rate,. — This salt crystallizes from alcohol or water in fine needles melting at 166 tt and decomposing just above this temperature.

Anal. Calcd. for CisH l6 0 3 H 2: 0, 62.90; H, 6.45; N, 11.29; Founc: 0, 63.19; H, 6.39; M, 11.30.

Fy dipchlo ride.—On dissolving the base in a slight excess of concentrated hydrochloric acid, the hydrochloride separates 1 X J

and can be recrystallized from alcohol or water in long slender needles melting at 212-214$ with decomposition.

Anal. Calcd. for 0, 70.40; H, 7.22; I, 6.32; 01, 16.02; Found: 0, 70.43; H, 7.24; B, 6.33; 01, 16.06

-cid Sulfate. —This salt, prepared by adding a slight excess of concentrated sulfuric acid to an acetone solution of the base, crystallizes from alcohol in fine needles melting at 339-240$ undecomposed.

Anal, Calcd. for N, 4.95; Found: H, 4.98.

j ercuric Chloride Salt.—Mercuric chloride (one mole) added to the base (one mole) in dilute hydrochloric acid solution precipitates the double salt. It crystallizes from water in large needles and from alcohol in microscopic needles melting at 212-214$ undecomposed.

fnal. Calcd. for HCI •HgClgf I, 2.84; Found: H,'2.88.

Chromate.—On addition of potassium chromate to the hyorochloride in weak acid solution, the chromate separates in a gelatinous form which on standing becomes crystalline. It crystallizes from glacial acetic acid in microscopic rhombic prisms which decompose around 100%

Anal. Cal cd. for K ’ 4 - 62 > Found: 8, 4.61. 2,3- Dimethylquinoline-8-Carboxylic Acid.—l mixture of 3 g. of 2,3-dimethyl-8-ethylquinoline, 6 g. of chromic anhydride and 78 co. of 1:5 sulfuric acid is refluxed for 60 hours. The solution, after neutralization with ammonium hydroxide to where precipitation of chromic hydroxide begins, is extracted with chloroform in a yield of about 190. For purification, the reaction product is decolorized in aqueous solution with flit* char, followed by recrystallization from alcohol and water in long needles melting at 202*. A mixed melting point with an authentic sample of 2,3-dimethylquinoline-8-carboxylic

showed no depression.

Anal. Oalcd. for C. 71.64; H, 5.47; N, 6.97; Found: 0, 71.64; H, 5.46; N, 6.87.

3,3-Dimethylqu^. — The above acid was decarboxylated by distillation with soda-lime and the resulting base was dissolved. in ether and precipitated with an alcoholic solution of picric acid. The picrate recrystallizes from glacial acetic acid in microscopic hexagonal prisms melting at 231°. This product, mixed with an authentic sample of 2,3-dimethylquinoline picrate, showed no depression in melting point.

u-M ethyl-8-ethyl-3-dim ethy l olmet hyl quinolin e. —ln conformity with the procedure of Koenigs and Stockhausen, - 5 g.

of bhe dimethylethylquinoline and 25 ml. of formalin in a sealed tube are heated on a steam cone for 45 hours. The reaction mixture, processed in the usual way, yields the dimethylol compound which, after recrystallization from chloroform and ethyl acetate, melts at 94-95°.

Anal. Calcd. for C l 5 C, 73.47; H, 7.75; H, 5,71; Found: C, 73.31; H, 7.69; 5.91.

Hcrn.tj.—This salt precipitates on the addition of an alcoholic solution of picric acid to the dimethylol base in ether and can be recrystallized from water or alcohol in long slender needles melting at 165.5*.

Anal. Calcd. for 0, 53.16; H, 4.64; W, 11.81; Found: C, 53.47; H, 4.80; 5, 11.88.

3-lethylsB-ethylquinoline.—The dimethylol compound (5 g.) in 250 ml. of 1:1 nitric acid is boiled for 6 hours. On evaporation of the solution, a practically quantitative yield of 3-methyl-8-ethylquinoline-2-carboxylic acid melting at 84-85* results. Since a well known procedure is involved here for conversion of a methyl at position 2 to carboxyl in quinoline homologs, this product was not analyzed. After decarboxylation at a temperature of 180°, the resulting base is purified through recrystallization of its picrate (A) from benzene and glacial acetic acid. The assigned structure, 3-msthyl-8-ethylquinoline, was confirmed by synthesis.

Synt hes is of 3-met hyl-8-et hylquinolin e.—Dry hydrogen chloride is introduced at a moderate rate into a mixture of propionic aldehyde (5 g.) and methylal (6 g.) for 5 minutes. On the addition of o-ethylaniline (1.8 g.) in concentrated hydrochloric acid (5 g.), a reaction begins and is completed by boiling the solution, for 3 hours. An equal volume of water is added, the filtered solution is treated with a slight excess of sodium nitrite, and after concentration to a small volume, the bases are liberated with sodium hydroxide and extracted. with ether. Final purification of the methyl ethylquinoline, after fractional distillation at diminished pressure, is effected through the picrate (B). The identity of salts A and B was confirmed by a mixed melting point determination. The new base which was obtained in a yield of 3p has the following physical constants: b.p. 263* with partial decomposition at 746 mm.; n 25/D, 1.5946.

Picrate A. Anal. Calcd. for 0, 54.00; H, 4.00; H, 14.00; Found: 0, 53.84; H, 4.07; K, 14.02.

Jicrate B. Found: C, 54.01; H, 4.00; I, 14.03.

Synthesis of 2,3-dim ethyl-8-ethylquinoline.—Final confirmation of the structure of the original kero base was effected through its which can be carried out as fol-

lows: a mixture of concentrated hydrochloric acid (f moles), tiglic aldehyde (1 mole), and o-ethylaniline hydrochloride (4 moles) is heated for 4 hours on a water bath and then steam distilled. The solution, after liberation of the bases with sodium hydroxide, is agaih steam distilled and. the distillate, acidified with hydrochloric acid, is treated at room temperature with sodium nitrite. Following the removal of the supernatant oil by ether extraction, the solution is boiled, filtered, made alkaline with sodium hydroxide, steam distilled and again extracted with ether. The base, regained from the solvent, gave a picrate which, admixed with the kero base picrate, produced no depression of the melting point (320 e ).

anal. Calcd. for 0, 55.07; H, 4.35; M, 13.53; Found: C, 55.19; H, 4.27; N, 13.46.

£ ynt hes is_ of 2,3,4,8-t et ra met hyl qu i nol in e . —The synt hes i s of this base was effected by condensing methyl acetyl acetone, CH 3 OO(CF 3 )COCH § (.7 mole) and o-toluidine (.6 mole) in the presence of concentrated hydrochloric acid (1 mole) with refluxing for 8 hours at water bath temperature. This quinoline homolog had not been reported previously in the literature. Physical constants and analytical data will be published later.

Ferrin and Bailey: J. Am. Chern. Soc. , 55, 4136 (1933).

* These bases were processed further with water extractions; however the data are not included in this problem.

Bratton and Bailey: J. Am. Chern. &oc., 59, 175 (1937).

b 3 King and Bailey: op. cit., p. 1249.

Koenigs and Stockhausen: Ber., 34, 4331 (1901).

$$ Of. Carl Beyer: J. prakt. Chew. , 33, 419 (1886); von Miller and Kinkelin: Her., 20, 1916 (1887T7

bt Cf. Boehner and von Miller: ibid., 16, 3464 (1883).

57bO ml. bases b.p. 283-'287'® 5700 ml. 1:1 hydrochloric acid 5700 ml. chloroform 4900 ml. bases n25/D 1.5315 2000 ml. water 1700 ml. bases n25/D 1.5616 1200 ml. chloroform 75 ml. bases 1600 ml. bases n2S/D 1.5762 n25/D 1.5215 b.p. 280° 800 ml. chloroform 40 ml. bases 1550 ml. bases n25/D 1.5778 500 ml. clil o rof orm n25/D 1.5310 b.p. 281° 25 ml. bas es 1535 ml. baa es B25/D 1.5792 n35/3 1.5431 b.£. 281.5 s 40G nil. chloroform 10 ml. bases 1500 ml. bases n25/D 1.5812 n25/D 1.5438 b.p. 282° b.p. 384°

Table 5. Cumulative Extraction Data on 5.7 Liters of Bases from Fifth Distillation, b.p. 283-287°, n 25/D 1.5350 to 1.5410

VJ N M W N N W W M M H H H H H H H H M H tO CD -O C" CT to tv H 8 O CO 00 to CD CH to to H» O <0 CO to tfx to to M <rl" f c+ co Ui w on cn cn cn to to to. tn ui ui on tn ui to to ui to to on to on cn on to on ui to OOOOOOOOOOOOOOOOOOOOOOOOOOOOO ch erf* M w* |-4 d P c+ o ♦ El » co CO DO co CD co to w to co co co to to to co to to to co CO CO CO tv CO CO to to to CO 00 CO CO CO CO 03 CO CO CO CO 00 CO CO CO CD 00 CO to -O <J -0 to ■<? to -2 to -0 CP to to to G2 6> UI UI CO >>• O'? to to to H H 4 O CO CO to CP CT.' CH W W O CO 8 CD 02 cf 01 £ ■* ♦ ♦ • ♦ • • H un ui ui tn to ui m ♦ CD o H H H F H H H H H H H i- 4 H H H H F 4 H H H H H H H H H H H rs to to to Ul to to to Ul to to to Ul to to to Ul to to to to to Ul to to to to to to U! 00 (X) GO CO 03 03 CO 00 00 03 GO CD CD 00 CO CO CO to to to to to to to to to CD CH GO co Ul to W to UI to to to i> to to to F 4 H O CD CO CP CD CD Ul H CO CO CO t)

Table 6. Distillation Data on 1500 co. Bases (Aromatic) 285° Fraction

Distillation Pressure 5 to 7 mm. Out # Bath T. Still Head < ml. Ata.b.p. e c. T. ®C. Obtained ■®c. 1 110 60 34 2 115 65 34 275 3 120 70 34 275 4 122 75 34 276 5 125 80 34 278 6 125 80 34 278.5 7 128 85 34 279 8 130 87 34 279 9 130 88 34 279.5 10 130 89 34 279.5 11 132 90 34 280 13 134 90 34 280 13 135 92 34 280.5 14 135 94 34 281 15 138 95 34 281 16 140 95 34 281.5 17 140 96 34 281.5 18 145 98 34 282 19 145 98 34 282 20 148 100 34 282.5 21 150 101 34 282.5 22 155 102 o 4 283 23 155 103 34 283 24 158 105 34 283.5 25 160 107 34 284 26 160 108 34 284 27 162 109 34 285 28 165 111 34 285.5 29 170 118 34 286 SO 180 130 34 31 190 145 34

Table 7. Amplified Distillation Data Obtained by Using 50 ml. Aromatic Bases b.p. 283.5°

CONCLUSION

The isolation of 2,3-dim ethyl-8-ethylquinoline (l) from the complex mixture of bases, obtained, from the kerosene distillate of California asphalt-base petroleum is described. The structure assigned this product has been confirmed by synthesis.

2,3-Dimethylquinoline (II) and 3-methyl-8-ethylquinoline (III) were obtained by indirect dealkylation of base I at positions 8 and 2, respectively. Base II is one of the kero quinolines which were isolated in the Texas Laboratory prior to the present investigation. Base 111 being a new quinoline homolog, its structure was established by synthesis.

BIBLIOGRAPHY

Ahrens, F. B.: Ber., 28, 745 (1895); 29, 2996 (1896).

Ahrens, F. B. and Gorkow, R.: ibid., 37, 2062 (1904).

Armenct, B.F. and Bailey, J. H.: J. Am. Chew. Soc., J>s, 4145 (1933).

Axe, W. N.: Unpublished work, University of Texas.

Axe, KN. , Henson, D. D. and Schuhardt, V. T.: I n d. Eng. Chem., 29, (1937).

Biggs, B. S.: Dissertation, University of Texas (1933).

Biggs. B. S, and Bailey, J. R. : J. Am. Chern. Soc., 55 4141 (1933).

Blicke, F. F. and Powers, J. L.: Ind. Eng. Chem., 19. 1334 (1927).

Bratton, A. C.: Dissertation, University of Texas (1936).

Bratton, A. C. and Bailey, J. R.: J. Am. Chem. Soc., 59, 175 (1937).

Bratton, A. C., Felsing, W. A. and Bailey, J. R.: Ind. Eng. Chem., 28, 424 (1936).

von Braun, J.: Ber., 57, 387 (1924).

Brown, R. L. and Cooper, R. B.: Ind. Eng. Chem., 19, 26 (1927).

Dav, D. T.: Handbook of the Petroleum Industry, John VTiley and ~

Dinwiddie, J. A,: Thesis, University of Texas (1935).

Eguchi, T.: Bull. Chern. Soc. Japan, 2, 176 (1927); 3, 227, 233 (1928).

Engler, C.: Chern. Ztg. , 30, 715 (1906).

Ganguli, S. K. and Guha, P. C.: J. Indian Chern. Soc., 11, 197 (1934).

Hahn, A.: Her., 43, 419 (1910).

Heap, J. G., Jones, W. J. and Speakman, J. B.: J. Am. Che®. Soc., 43, 1936 (1921).

Jacobsen, E. and Reimer, 0. L.: Ber., 16, 297, 2606 (1883); 19, 2427 (1886).

Johnson, T. B. and Daschavsky, P. G.: J. Biol. Chern., 62, 197 (1924-5); J. Ab. Ghem. Soc., 41, 1147 (1919).

King, W. A. and Bailey, J. R.: J. Am. Chern. Soc., 52, 1239, 1249 (1930).

Koenigs, W. and Stockhausen, F.: Ber., 34, 433 (1901).

Koenigs, W. and Nef, I.: ibid., 19, 2437 (1886).

Lackey, R. Dissertation, University of Texas (1934).

Lackey, R. W. and Bailey, J. R.: J. Am. Chea. Soc., 56, 2741 (1934).

Lake, G. R.: Thesis, University of Texas (1933).

Lake, G. R. and Bailev, J. R.: J. Am. Chem. Soc., 55, 4155 (1933).

Mabery, C. F.: J. Soc. Chem. Ind., 19, 505 (1900).

Mabery, C. F. and Wesson, L. G.: J. Arc. Chem. Soc. , 43, 1016 (1920).

Mahan, R. I.: Unpublished work, University of Texas.

Mahan, R. I.: Thesis, University of Texas (1935).

McKee. R. H.: Shale Oil, Chemical Catalog Co., Inc., New York, (1925).

Meadows, J. 1.: Unpublished work, University of Texas.

Mchelmnn, J.: Ind. Eng. Chern., 17, 471 (1925).

Oparina, M. P.: Ber., 648. 566 (1931).

Parker, Ivy 1.: Dissertation, University of Texas (1935).

Parker, Ivy IL, Gutsseit, 0. L., Bratton, A. 0. and Bailey, J. R.: J. Am. Chern. S oc ., 58, 1097 (1936).

Parker, Ivy M. and Bailey, J. R. : ibid., p. 1102.

Perrin, T. S.: Dissertation, University of Texas (1933).

Perrin, T. S. and Bailey, J. R.: J. Am. Chem. Soo., 55, 4136 (1933). ~

Peters, W. a. and Baker, T.: Ind. Eng. Chem., 18, 69 (1926).

Porter, E. C.: Coal Carbonization, Chemical Catalog Co., Inc., lew York (1924>.

Poth, E. J., Armstrong, W. D Cogburn, C. C. and Bailey. J. R.: Ind. Eng. Chem., 20, 83 (1928).

Poth, E. J. and Bailey, J. R.: J. Am. Chern. Soc. , 52, 1249 (1930). ~

Podbielniak, W. J.: Ind. Eng. Chem., Anal. Ed., 3, 177 (1931).

Pyhala, E.: Chern. Ztg. , 46, 933 (1922).

Schiff, H.: Ann., 140, 125 (1866).

Spielman, P. E.: Constituents of Coal Tar, Chemical Catalog Co., Inc., Kew r York (19&&J, pp. 163-192.

Thompson, W. G.: Dissertation, University of Texas (1930).

Thompson, C. and Bailey, J. R.: J. Am. Chem. Soc. , 53, 1002 (1931).