US2510875A - Recovery of quinaldine - Google Patents

Description

Patented June 6, 1950 RECOVERY OF QUINALDINE Karl Henry Engel, Teaneck, N. J assignor to Allied Chemical & Dye Corporation, New York, N. Y., a corporation of New York No Drawing. Application February 16,1946, Serial No. 648,209

6 Claims.

This invention relates to the separation of quinaldine from mixtures of this base with other nitrogen bases.

Quinaldine is customarily extracted along with other nitrogen bases from coke oven distillates and other sources of heterocyclic nitrogen bases by means of an aqueous solution of a mineral acid, usually sulfuric acid. After liberation of the bases from these acid solutions by means of alkali, the base mixtures obtained are customarily fractionally distilled to obtain fractions predominating in a particular base. Quinaldine fractions may thus be obtained boiling predominantly within the range of about 245 to 250 C. and containing amounts of quinaldine ranging from a minor percentage to 80% or more, but even closely cut fractions will contain substantial amounts, e. g. 20% or more, of other nitrogen bases having boiling points close to quinaldine and not readily separable from the quinaldine by distillation, e. g. quinoline, isoquinoline and the various methyl quinolines and methyl isoquinolines. Mixtures of this kind have found limited practical application and are of relatively low economic value. Pure quinaldine, however, is a product of considerable commercial interest, being useful, for example, in the preparation of pharmaceutical products and dyes. The purity requirements in such fields may be quite rigid.

As above indicated, the bases present with quinaldine in its commercial fractions are not readily separable from the quinaldine by fractional distillation. No description of a practical method for separating these closely related compounds has been found in the published prior art. In the case of various other base mixtures, methods usually depending upon fractional crystallization of the less common salts of the bases present, e. g. picrates, ferricyanides, platinichlorides, etc., have been proposed. No suggestion has been found, however, that these or any other reagents are suitable for separating substantially pure quinaldine from like-boiling bases.

It is an object of this invention to provide a practical and economical method for separating pure quinaldine from mixtures thereof with closely related nitrogen bases not readily separable therefrom by distillation.

It is a further object of this invention to provide a commercially feasible process for isolating pure quinaldine or fractions enriched in quinaldine from the usual commercial quinaldine fractions, particularly tar base fractions boiling predominantly in the range 245 to 250 C.

I have discovered that quinaldine of 95% to 100% purity may be preferentially precipitated as hydrochloride from mixtures of quinaldine with isoquinaline and other closely related nitrogen" bases not readily separable from quinaldine by distillation, provided the weight ratio of isoquinoline to quinaldine in such mixture is not more than 2:1, preferably not more than 1:2, by treating such a base mixture in a substantially anhydrous medium with a limited amount of hydrogen chloride or equivalent reagent for the formation of base hydrochloride.

Such preferential precipitation of quinaldine hydrochloride to the substantial exclusion of other base hydrochlorides sets quinaldine apart from such closely related bases as quinoline, for example, which is not precipitated in the form of its hydrochloride even from fractions containing it in greatly predominating amount. This preferential precipitation also distinguishes hydrogen chloride, for this particular reaction, from other salt-forming reagents such as sulfuric acid, which I have found are ineffective to precipitate a base such as quinaldine selectively. Reasons for the preferential precipitation of quinaldine hydrochloride are not fully understood but appear to depend on phase rule relations which are unpredictable and which in this case are complex because of the number of the substances involved. The relative basicity of the bases is not a factor when operating in accordance with my invention as indicated by the fact that the precipitate does not contain the various bases of the original base mixture in amounts which would be expected from the we1l-known distribution laws.

A suitable quinaldine crude for use in the process of my invention may advantageously be obtained by fractional distillation of the quinaldinecontaining base mixtures recovered from the usual sources of such mixtures, for example, coal tar. It has been found that quinaldine-containing base mixtures boiling for the most part within the range 245 to 250 C. have suitable properties for precipitation of substantially pure quinaldine in the form of its hydrochloride. The quinaldine fractions of this type generally obtainable usually contain from about 35% to about 80% quinaldine and an amount of isoquinoline no more than about twice the weight percent of the quinaldine, the remainder consisting of quinoline, methyl quinclines, methyl isoquinolines, and other nitrogen bases of like boiling point. A close boiling fraction of boiling range 245 to 247.5 C. containing from to quinaldine and less than 10% isoquinoline is the preferred material for use in the process of my invention. a

The liquid medium for carrying out precipitation of quinaldine hydrochloride may be merely the non-reacting tar bases in the mixture being treated or may be such tar bases augmented by an added diluent such as a hydrocarbon or chlorinated hydrocarbon. The precipitation should be carried out in the absence of water, alcohols, e. g. methanol, ethanol or propanol, and dioxanes. As pointed out below, when water is introduced into the reaction mixture, for example, with the reactants, it should be removed, e. g. by azeotropic distillation as more fully described below, in order to bring about precipitation of the desired base hydrochloride.

In a preferred method of carrying out the process of my invention, a commercial quinaldine fraction as above described is treated either with gaseous hydrogen chloride or with aqueous hydrochloric acid which may advantageously be the inexpensive and readily available technical grade of muriatic acid. This treatment is preferably carried out in the presence of a diluent as above discussed, e. g. an inexpensive hydrocarbon solvent such as benzene, toluene, xylene, a petroleum solvent, or a naphthenic hydrocarbon such as oyclohexane or methylated cyclohexanes.

In order to obtain an anhydrous crystallization medium when an aqueous hydrochloric acid is employed, the mixture is heated to refluxing temperature to remove the water by azeotropic distillation, the free nitrogen base and added diluent, if present, acting as azeotropic agent. The dis tillate is condensed in a side-arm trap where it separates into a water layer and an oil layer. The water layer is removed while the oil layer containing unreacted base and diluent, when such has been used, is continuously returned to the still. When all or nearly all of the water has thus been removed, the still charge is cooled to bring about or to complete crystallization of quinaldine hydrochloride.

In general, the desired base hydrochloride precipitate may be formed at the normally prevailing temperature; i. e. cooling below room temperature is not usually necessary.

As above stated, a limited amount of hydrogen chloride or equivalent reagent is employ :1. By employing relatively small amounts or" acid reagent, substantially pure quinaldine hydrochloride may be precipitated, i. e. of hydrogen chloride up to the entity that would be required in theory to react quantit of nitrogen base equivalent t.-.) quinaldine present minus one-half the isoqrinoline present, may be employed.

The precipitated crystals of base hydrochloride are isolated b filtration, e. g. centrifuging, and are freed from adhering unreacted bases by washing with a suitable solvent, e. g. the solvent used as diluent for the precipitation. The Crystals are then freed from adhering solvent by drying or other means. Quinaldine is liberated from the crystals thus obtained, for example, by adding alkali to an aqueous solution of the quinaldine hydrochloride. The liberated base, which separates out from the resulting salt solution, is dehydrated (for example, by treating with solid caustic soda) and. preferably distilled to remove traces of colored impurities.

The mother and wash liquors may be distilled to recover solvent for reuse in the'process and the residual tar bases may then be fractionated to produce fractions rich other bases, e, g. isoquinoline, quinoline, etc.

The following examples are illustrative of the process of my invention:

Example 1.-A quinaldine fraction, boiling range 245.0" to 247.5 0., containing about 79% quinaldine, about 7% isoquinoline and 14% other bases such as quinoline and various methyl quinolines and methyl isoquinolines, was used as a source of quinaldine.

The apparatus employed consisted of an acidresistarrce reaction vessel equipped for agitation, a jacket suitable for heating with steam or for cooling with circulating tap water, and a reflux condenser. Distillation condensate was not immediately returned to the vessel but was passed through a settling trap which permitted a separation of mixed distillates consisting of aqueous and oil layers of liquid. The trap was arranged for withdrawal of the aqueous layer and return of the oil layer to the reaction vessel.

This vessel was charged with 700 parts by Weight of the crude quinaldine fraction diluted with 70 parts by weight of a petroleum naphtha fraction of approximate boiling range to C. The charge was partly neutralized by gradual addition of 160 parts by weight of concentrated (35.5%) hydrochloric acid. The mixe ture was heated to refluxing and condensate was allowed to separate into two layers, the upper oil layer being continuously returned to the stiil. lower water layer was removed until 126 by weight of water had been colle ed, at which time the still charge was practically dchydrated. The temperature inside the flask rose from 35 to 125 C, during dehydration.

At the end of the dehydration, a heavy slurry of crystals had formed. The temperature of the charge was raised to C. so as to remove some of the petroleum solvent and redi mlve most of the precipitate. The melt was poured into 350 parts by weight of petroleum solvent with agitation to produce a uniform crystal slurry. Crystals were filtered, washed with solvent, and dried at 70 C. A yield of 268 parts by Weight of a product of melting point 224 C. was obt med- These crystals of quinaldine hydrochloride were dissolved in 500 parts by weight of water, the solution was slightly evaporated to remove last traces of solvent, and quinaldine was liberated by addition of a solution of 80 parts by weight of sodium hydroxide in 100 parts weight of water. A yield of 205 parts by weight of dried, distilled quinaldine was obtained. The melting point of 3.l C. indicated a purity of 9.7%.

Example 2.In a similar run, 500 parts by weight of the quinaldine fraction, 70 parts by weight of the petroleum naphtha, and 229 parts by weight of 35.5% hydrochloric acid were were charged to the dehydration apparatus. The initial lumpy precipitate that formed when dehydration was complete was largely dissolved by heating the charge to C. and the melt was then poured into 280 parts by weight of petroleum solvent. The yield of quinaldine hydrolchloride crystals was 374 parts by weight, melt ing point 212 C,

These crystals were dissolved in 500 parts by Weight of water, the solution was filtered to remove a impurity which appeared as a, colloidal precipitate at this point, and quinaldine was liberated from the filtrate by treatment with 100 parts by weight of sodium hydroxide in water soe lutign. The quinaldine was dried with sodium hydroxide and distilled. A yield of 268 parts by weight (68% of the quinaldine in the charge) of quinaldine of limpid point 5.0 C. (the temperature at which a suspended solid phase just disappears upon heating) indicated a quinaldine content of 95.8%.

Since certain changes may be made in carrying out the above process without departing from the scope of the invention, it is intended that all matter contained in the above description shall be interpreted as illustrative and not in a limiting sense.

The term hydrogen chloride is employed in the claims to include anhydrous hydrogen chloride and aqueous hydrochloric acid.

I claim:

1. In a process for separating an enriched quinaldine product from a mixture of quinaldine with like-boiling nitrogen bases in which the weight ratio of quinaldine to isoquinoline is at least 1:2, the steps that comprise treating the mixture in a substantially anhydrous medium with hydrogen chloride in amount not greater than that theoretically required to react with all the quinaldine present whereby quinaldine hydrochloride is precipitated preferentially to other nitrogen base hydrochlorides, and separating this precipitate.

2. In a process for separating an enriched quinaldine product from a mixture of quinaldine with like-boiling nitrogen bases in which the weight ratio of quinaldine to isoquinoline is at least 2:1, the steps that comprise treating the mixture in a substantially anhydrous medium with hydrogen chloride in amount not greater than that theoretically required to react with all the quinaldine present whereby quinaldine hydrochloride is precipitated preferentially to other nitrogen base hydrochlorides, and separating this precipitate.

3. In a process for separating an enriched quinaldine product from a mixture of quinaldine with like-boiling nitrogen bases in which the weight ratio of quinaldine to isoquinoline is at least 1:2, said mixture boiling predominantly in the range 245 to 250 C., the steps that comprise treating the mixture in a substantially anhydrous medium with hydrogen chloride in amount not greater than that required in theory to react with the quinaldine present whereby quinaldine hydrochloride is precipitated preferentially to other nitrogen base hydrochlorides, and separating this precipitate.

4. In a process for separating substantially pure quinaldine from a mixture of quinaldine with like-boiling nitrogen bases in which the weight ratio of quinaldine to isoquinoline is at least 2:1, said mixture boiling predominantly in the range 245 to 247.5 0., the steps that comprise treating the mixture, in an anhydrous medium and in the absence of alcohols, with hydrogen chloride in amount not greater than that required in theory to react with an amount of nitrogen base equivalent to the quantity of quinaldine present minus one-half the isoquinoline present, to precipitate substantially pure quinaldine hydrochloride preferentially to other nitrogen base hydrochlorides, and separating this precipitate.

5. In a process for separating substantially pure quinaldine from a mixture of quinaldine with like-boiling bases in which the weight ratio of quinaldine to isoquinoline is at least 1:2, the steps that comprise treating the mixture with aqueous hydrochloric acid in amount limited to react in theory with substantially less than the total quinaldine present, subjecting the mixture to azeotropic distillation to remove water and cause quinaldine hydrochloride to be precipitated preferentially to other nitrogen base hydrochlorides, and separating this precipitate.

6. In a process for separating substantially pure quinaldine from a mixture of quinaldine with like-boiling nitrogen bases and containing at least about of quinaldine, said mixture boiling predominantly in the range 245 to 247.5 0., the steps that comprise treating the mixture in an anhydrous medium and in the absence of alcohols with hydrogen chloride in an amount not greater than that required in theory to react with an amount of nitrogen base equivalent to the quantity of quinaldine present minus onehalf the quantity by weight of the isoquinoline present to precipitate substantially pure quinaldine hydrochloride preferentially to other nitro gen base hydrochlorides, and separating this precipitate.

KARL HENRY ENGEL.

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

UNITED STATES PATENTS Number Name Date 1,965,828 Fox July 10, 1934 2,035,583 Bailey Mar. 31, 1936 2,288,281 Huijser et al June 10, 1942 2,309,324 MoAllister Jan. 26, 1943 2,432,064 Cislak et a1 Dec. 2, 1947 OTHER REFERENCES Heller et a1.: Berichte 41, page 2701 (1908).

Claims (1)

1. 4. IN A PROCESS FOR SEPARATING SUBSTANTIALLY PURE QUINALDINE FROM A MIXTURE OF QUINALDINE WITH LIKE-BOILING NITROGEN BASES IN WHICH THE WEIGHT RATIO OF QUINALDINE TO ISOQUINOLINE IS AT LEAST 2:1, SAID MIXTURE BOILING PREDOMINANTLY IN THE RANGE 245* TO 247.5*C., THE STEPS THAT COMPRISE TREATING THE MIXTURE, IN AN ANHYDROUS MEDIUM AND IN THE ABSENCE OF ALCOHOLS, WITH HYDROGEN CHLORIDE IN AMOUNT NOT GREATER THAN THAT REQUIRED IN THEORY TO REACT WITH AN AMOUNT OF NITROGEN BASE EQUIVALENT TO THE QUANTITY OF QUINALDINE PRESENT MINUS ONE-HALF THE ISOQUINOLINE PRESENT, TO PRECIPITATE SUBSTANTIALLY PURE QUINALDINE HYDROCHLORIDE PREFERENTIALLY TO OTHER NITROGEN BASE HYDROCHLORIDES, AND SEPARATING THIS PRECIPITATE.