PREPARATION OF SECONDARY AMINES. 2433CCL-Pwparation of Secondary Amines from Cuq*b-oxylic Acids. Pa9-t I. P~re~xwatio~~ of Hepta-decylaniline, Pentaclecylaniline, ayhd Fridecyl-aniline.By HENRY RONDEL LE SUEUR.THE method most generally employed for the preparation ofsecondary amines consists in the interaction of an alkyl haloidcompound and a primary amine. A serious objection to this methodis that tertiary amines are also formed, and the subsequen2434 LE SUEGR : PREPARATION OF SECONDARYseparakion of the secondary amines from these is a matter of con-siderable difficulty. Further, this method necessitates the use ofalkyl monohaloid compounds, and these, with the exception of thelower members, are not always easy to prepare.The method which is now brought forward for the preparation ofmonoalkylanilines, and which, so far as can be ascertained, is new,is free from these two objections, as, firstly, owing to the natureof the reaction, there is no possibility of the formation of dialkyl-anilines, and, secondly, the entering alkyl group results directlyfrom an acid, and acids are more easily obt.ained,than any otherclass of organic compounds.The new method is briefly as follows:(1) the a-anilino-acid is prepared by the interaction of aniline andthe a-bromo-acid, ('2) the a-aniIinobacid is heated to considerablyabove its melting point, whereby it loses carbon dioxide, and amonoalkylaniline results :CH,*[CH,],,*CHBr*CO,H + C6H,*NH, =a-Bromostearic acid.CH3-[CH2],,*CH(NHPh}*C02H + HBr.a-Anilinostearic acid.CH3*[CH,],,*CH(NHPh)*C02H Et",d CH,-[CH,],,*CH,*NIIPh + CO,.I n the three caes so far investigated, and which form the subjectof this communication, the yield of a-anilineacid was more than70 per cent.of the theoretical, and the yield of alkylaniline wasalso more than 70 per cent. of that theoretically obtainable fromthe anilino-acid.a-Anilinostearic acid, a-anilinopalmitic acid, and a-anilinomyristicacid have been already prepared by Hell and his collaborators (Rer.,1889, 22,1748; 1891,24, 942,2395), who obtained these compoundsby the interaction of aniline and the respective a-bromo-acid a t180-185O. Heating to 180-185° is not necessary, and is to beavoided, as at this high temperature there is a likelihood of theanilino-acid undergoing decomposition, and also of the formation ofan anilide and of an aS-unsaturated acid, the latter being formedby the removal of hydrogen bromide from the a-bromeacid by theaniline.Heptadecylaniline, C',7H,,*NH*C6H,, pentadecyluniline,C,,H,,*NH-C,H,, and tridecylanilime, Cl3H2,-NH=C,H,, are colour-less solids, which melt at a low temperature, and are readily solublein most of the common organic solvents. Their hydrochlorides areinsoluble in cold water, and when heated with this liquid they meltand undergo almost complete hydrolysis into the free base andhydrogen chloride.The author is now engaged in the investigation of the applicationof the above reaction to the preparation of secondary amines inHeptadecylanilineAMINES FROM CARBOXYLlC ACIDS.PART 1. 2435general, and more particularly of monoalkylnaphthylamines andother monoalkylanilines, and from the results so far obtained it isevident that this method is not limited to the preparation of thehigher monoalkylanilines.EXPERIMENTAL.Preparation of a-Anilinostearic Acid, ClGH,,*CH(NHPh)*CO2H.Thirty grams of a-bromostearic acid (1 mol.) and an equal weightof aniline (4 mols.) were heated together in a flask immersed inboiling water for thirteen hours, The resulting solid was thoroughlydigested with excess of hot dilute hydrochloric acid, collected, andwashed with dilute acid and water to remove excess of aniline; i twas then dried and cryst'allised from a mixture of alcohol and ethylacetate, when 23 grams of the pure acid were obtained.(Found,C = 76.54 ; H = 10.75 ; N = 3-91. Calc., C = 76-80 ; H = 10'93 ;N=3*73 per cent.)a-Anilinodearic acid is sparingly soluble in alcohol, acetone,benzene, or chloroform in the cold, but dissolves readily on heating.It is insoluble in water, ether, or light petroleum, and separatesfrom ethyl acetate in nodular aggregates, melting at 141--142O, andnot at 84*5O, as stated by Hell and Sadomsky (Ber., 1891, 24,2395). The low melting point given by Hell and Sadomsky isobviously not in agreement with the value to be expected fromanalogy to other similar a-anilino-acids.Heptadecylaiziline, Cl7H,,*NH*C6H,.Five grams of a-anilinostearic acid were placed in a small flaskcontaining a thermometer, the bulb of which dipped into the sub-stance, and the whole heated in a metal-bath.As soon as thesubstance was melted, the temperature was raised rapidly to about190°, at which point the evolution of carbon dioxide commenced.The temperature was then raised more slowly to 270-280°, andthe heating stopped when the evolution of carbon dioxide hadceased. The evolution of carbon dioxide is very rapid a t about220°, and there is no evidence of charring or secondary decom-position at any sta,ge of the heating, which for 5 grams of acidrequires about fifteen minutes. %'he product resulting from 20grams of anilineacid, heated in quantities of 5 grams at one timeas described above, was distilled under 50 mm. pressure, when 14-3grams of distillate, boiling between 295O and 300°, were obtained.This was redistilled under 35 mm.pressure, and gave:Below 285" ................................................ 2.3 grams.Undistilled residue = 2-0 grams.285-290" ................................................ 10.0 ,2436 LE SUEUR : PREPARATION OF SECONDARYThe fraction 285--290° readily solidified to a white solid meltingat 4 1 - - - 4 2 O , arid consisted therefore of the pure amine, and, asnearly the whole of the fraction distilled a t 285-286OY this tem-perature is to be regarded as the boiling point of the pure substance.The amine was also purified by crystallisation instead of fractionaldistillation, in which case the method adopted was as follows. T l eproduct resulting from the heating of the anilino-acid was distilledin a vacuum and the distillate dissolved in ether, the etherealsolution washed with a solution of potassium hydroxide, dried withsolid potassium hydroxide, and the residue left on evaporation ofthe ether crystallised from alcohol until its melting point wasconstant.This alternative method of purification is especiallyapplicable to the preparation of small quantities of the amine :0.1632 gave 0.5000 (20, and 0.1836 H,O. C= 83.55 ; H='12.50.0.2380 ,, 8.7 C.C. N, (moist) at 12O and 770 mm, N=4.40.C,,H,,N requires C = 83-38 ; H = 12.38 ; N = 4.23 per cent.Heptadecylmdine is readily soluble in ether, benzene, chloroform,acetone, light petroleum, or ethyl acetate in the cold, sparingly so incold alcohol, but readily so on warming, and crystallises from thissolvent.in large plates, which soon change to long needles, the latterbeing the stable crystalline form. Heptadecylaniline melts at42--43O, and boils at 285-286Oj35 mm. It is insoluble in hydro-chloric acid, but dissolves readily in concentrated sulphuric acid.The hydrochloride, CI7H3,*NH*C,H,,HC1, was obtained by dis-solving 1.5 grams of the amine in 50 c . ~ . of ether, and passing dryhydrogen chloride into the solution until saturated. The pre-cipitated hydrochloride was collected, wa*shed with ether, dried, andcrystallised from light petroleum (b. p. 60-80°), when it wasobtained in beautiful, thin, glistening plates, melting at 99-looo.It is sparingly soluble in alcohol, ether, acetone, benzene, or lightpetroleum in the cold, and readily dissolves in cold chloroform orboiling light petroleum.It is insoluble in cold water; in hot water,however, the substance melts, but does not dissolve, and the aqueousliquid acquires a strongly acid reaction. 0.2856 Gram, suspendedin hot water, required 7.8 C.C. ,V/lO-NaOH for neutralisation of theaqueous solution, using methyl-orange as indicator, whereas the sameweight of the compound, C17H35*NH*CGH5,HC1, requires 7.8 C.C.A'/ 10-NaOH.The acetyl derivative, C,,H3,*N(CH3*CO)*C,H5, was readily pre-pared by boiling together half a gram of the amine and two gramsof acetic anhydride for one and a-half hours. The excess of aceticanhydride was removed by allowing the product to remain in avacuum over a saturated solution of potassium hydroxide, and thAMINES FROM CARBOXYLIC ACIDS.PART I. 2437solid residue was purified by crystallisation from methyl alcoholcontaining a small quantity of water:0.1690 gave 5.9 C.C. N, (moist) at 20° and 750 mm.C,,H,30N requires N = 3.75 per cent.A cetoheptadecylanilide is readily soluble in most of the ordinaryorganic solvents, and crystallises from methyl alcohol containing alittle water in feathery aggregates of slender needles, melt,ing at42-43O.The nitrosoarnine, C17H3j*N(NO)*C6H6, was most easily obtainedby the following method, which gave a practically theoretical yieldof the pure substance. 1-5 Grams of the amine were dissolved in10 C.C. of concentrated sulphuric acid, and this solution added dropby drop to a solution of 3 grams of sodium nitrite in 80 C.C.ofwater, the whole being vigorously shaken, and more (about 2 grams)sodium nitrite added from time to time. The resulting solid wascollected, washed, dried, and crystallised from methyl alcohol :N = 3.94.0.1520 gave 10.5 C.C. N, (moist) at 18.5O and 760 mm.C,,H,,ON, requires N = 7.78 per cent.P~emyl~eptadecyllzitrosoamine is readily soluble in ether, chloro-form, benzene, light petroleum, or acetone in the cold, sparinglyso in cold methyl alcohol, but dissolves readily on heating, andcrystallises from this solvent in light f awn-coloured plates, meltingat 53-54O. A small quantity of the nitrosoarnine, mixed with alittle phenol and warmed with concentrated sulphuric acid, gave ablue solution, which, on dilution, gave a red, opalescent liquid,turning blue on being rendered alkaline.N= 7-96.Preparation of a-Anilinopdmitic Acid.Thirty grams (1 mol.) of a-bromopalmitic acid and an equalweight (3+ mols.) of aniline were heated togejher in a flask immersedin boiling water for seven hours, and the resulting product w5~9worked up as described for the preparation of a-anilinostearic acid(p.2435). The crude acid was purified by crystallisation from itmixture of alcohol and ethyl acetate, and was obtained in nodules,melting at 143--144O, a melting point which agrees with that givenfor this substance by Hell and Jordanoff (Ber., 1891, 24, 942). Theyield of pure acid obtained corresponded with 80 per cent. of thetheoretical.(Found, N = 4.09. Calc., N = 4.03 per cent.)Pemtadecylan&%ze, C,,H3,-NH*C6H,.The a-anilinopalmitic acid was heated in quantities of 5 grams ina flask immersed in a metal-bath, as described for the preparation ofVOL. XCVII, 7 2438 LE SUEUR : PREPARATLON OF SECONDARYheptadecylaniline (p. 2435). The evolution of carbon dioxide com-menced at about 190°, was rapid at 220°, and had ceased afterfifteen minutes' heating, the temperature having risen at the endof that time to 2 8 0 O . The product resulting from 29 grams ofanilino-acid heated as described above was distilled under 40 mm.pressure, when the following fractions were obtained :Below 274" ............ 2.8 grams. 1 290-340" .............. 2-3 grams.The fraction 274-290° was redistilled under 40 mm.pressure,when 16 grams, boiling at 271-27407 were obtained, which solidifiedto a colourless solid, melting at 32-33O, and consisted of the pureamine. The fraction boiling below 274O on crystallisation fromalcohol gave 1.6 grams of the pure substance. The total weight ofpure amine obtained from 29 grams of anilino-acid was 17.6 grams,which corresponds with a 70 per cent. yield of the theoretical:0.1488 gave 0.4548 CO, and 0.1640 H,O.0-2164 ,, 8.8 C.C. N, (moist) at 18O and 762 mm. N=4*71.C21H37N requires C = 83-17 ; H = 12-21 ; N =4.62 per cent.Pentadecylaniline is readily soluble in ether, benzene, chloroform,acetone, light petroleum, or ethyl acetate in the cold, is sparinglysoluble in cold alcohol, but dissolves readily on warming, andcrystallises from this solvent in feathery aggregates, melting at;34-35O.It is insoluble in hydrochloric acid, but dissolves readilyin concentrated sulphuric acid. The pure amine boils at 271°/40 mm.The hgdrochloride, C',,H,,*NH*C,H,,HCl, was readily obtainedby passing dry hydrogen chloride into a solution of 1 gram ofpentadecylaniline in 30 C.C. of ether, when the pure hydrochloridesoon crystallised in glistening plates, which melted at 97'5O. It isinsoluble in ether, acetone, benzene, or light petroleum in the cold,dissolves readily in cold chloroform and in boiling light petroleum,from which it separates in glistening plates on cooling. It isinsoluble in water, and when heated with this solvent it melts butdoes not dissolve, the water acquiring a strongly acid reaction.0.2574 Gram, suspended in hot water, required 7.50 C.C.N/10-NaOH for neutralisation, using methyl-orange as indicator, whereasthe same weight of the compound C,,H,,*NH*C,H,,HCl requires7.58 C.C. N / 10-NaOH.The acetyl derivative, C,,H,,=N(CH,*CO)*C,H,, was prepared byboiling 1 gram of the amine with 4 grams of acetic anhydride forfour hours. The excess of anhydride was removed by allowing theproduct t.0 remain in a vacuum over a concentrated solution ofpotassium hydroxide, and the residue crystallised from methylalcohol containing a very little water :274-290" ........... 17 '7 ,, I Untlistilled residue ... 3.0 ,,C = 83.35 ; H = 12.24AMINES FROM CARBOXYLIC ACIDS. PART I, 24390.2008 gave 7.5 C.C.N, (moist) at 27*5O and 766 mm.C,,H,,ON requires N = 4.06 per cent,Acetopentaclecylanilicle is very readily soluble in all the commonorganic solvents, and crystallises from methyl alcohol containing alittle water in hair-like needles, melting at 30*5-31'5°.The nitrosoamime, C,SH,,*N(NO)-C,H,, was prepared by addinga solution of 1 gram of the amine in 8 C.C. of concentrated sulphuricacid to a dilute solution of sodium nitrite, the latter being keptin considerable excess, as described for the preparation of thenitrosoarnine of heptadecylaniline (p. 2437). The crude nitroso-amine was purified by crystallisation from met.hyl alcohol :N=4*14.0.1420 gave 10.2 C.C. N2 (moist) at 16O and 768 mm.C21H,,0N2 requires N = 8.43 per cent.Phenylpentadecylnitrosoamine is readily soluble in ether, chloro-form, light petroleum, acetone, or benzene in the cold, and crystallisesfrom methyl alcohol in glistening, flat needles, which melt at 49O,and have a, light fawn colour.A small quantity of the nitrosoaminewarmed with phenol and concentrated sulphuric acid gave a deepblue solution, which, on dilution, gave a red, opalescent liquid,turning blue on being rendered alkaline.N = 8.46.Preparation of a-A niEino m yris tic A cid.Thirty grams (1 mol.) of a-bromomyristic acid and 32 grams(38 mols.) of aniline were heated together in a flask immersed inboiling water for eight hours. The product was worked up asdescribed for the preparation of a-anilinostearic acid (p. 2435), andthe acid purified by crystallisation from a mixture of alcohol andet'hyl acetate, from which it separated in nodular aggregates meltingat 142-143O (compare Hell and Twerdomedoff, Ber., 1889, 22,1748).(Found, N=4.79. Calc., N=4.39 per cent.)Tridecylaniline, C,3H27*NH*C6H,.The a-anilinomyristic acid was heated in quantities of 5 grams ina flask immersed in a metal-bath exactly as described for the pre-paration of heptadecylaniline and pentadecylaniline (pp. 2435,2437), the course of the decomposition being similar in all threecases. The product resuIting from t,he action of heat on 16 gramsof anilinoniyristic acid was distilled under 35 mm. pressure :240-245 "... .. . . . ,248-255 "... ...... 10.5 grams.The fraction 248-255O was redistilled under 35 mm.pressure,few drops only. 260-280" ... .. . . . . . . . ...Undistilled residue ... 1.7 ,, 1 *5 grams. 17 u 2440 PREPARATION OF SECONDARY AMINES.when 9.8 grams, boiling at 250-255O, were obtained, which solidifiedto long, flat needles, melting at 23-24O. This corresponds with a71 per cent. yield of the theoretical. A portion boiling at 251Owas collected separately for analysis :C= 82-65 ; H = 11.99. 0'1512 gave 0.4582 CO, and 0.1632 H,O.0'2052 ,, 9.6 C.C. N, (moist) at 15O and 766 mm. N=5-52.C19H3N requires C = 82-91 ; H = 12.00 ; N = 5-09 per cent.Tm'decylaniline is readily soluble in alcohol, ether, benzene,chloroform, acetone, or light petroleum, and crystallises fromrectified methyl alcohol in long needles, melting at 23-24O.Itis insoluble in water or hydrochloric acid, but dissolves readily inconcentrated sulphuric acid.The hydrochloride, C,3H27*NH*C6H,,HC1, was prepared by passingdry hydrogen chloride into a solution of 1.5 grams of the aminein 50 C.C. ether until saturated. Light petroleum was then addedto the ethereal solution, and the ether evaporated, when, on allowingthe resulting solution to cool, the hydrochloride separated inglistening thin plates. It is readily soluble in a.lcoho1, chloroform,or benzene in the cold, is insoluble in cold acetone or ether, andcrystallises from light petroleum (b. p. 60-80°) in glistening, thinplates, melting at 94'5-95'5O. When heated with water, it meltsbut does not dissolve, and the water becomes strongly acid.0.3060 Gram, suspended in hot water, required 9.8 C.C. N/10-NaOH, using methyl-orange as indicator, whereas this amount ofthe compound @13H2,*m*C6H,,HC1 requires 9.8 C.C. N / 10-NaOH.The acetyl derivative, C,3H2,*N(CH3*CO)DC6H,, was prepared bythe interaction of the amine and acetic anhydride, as described forthe preparation of the other acetyl derivatives :It boils at 251°/35 mm.0.1852 gave 7.6 C.C. N, (moist) at 1 5 O and 756 mm.C2,H3,0N requires N=4.41 per cent.&4 cetotridecylandide is readily soluble in all the common organicsolvents, and crystallises from its solution in dilute methyl alcoholwhen cooled in a mixture of ice and salt, in flat needles, melting atThe nitrosoarnine, C13H2,*N(NO)oC6H,, prepared in a mannersimilar to the other two nitrosoamines, was crystallised from methylalcohol :N=4*77.31-32'.0.1520 gave 12.5 C.C. N, (moist) at 19'5O and 769 mm.C,gH3,0N2 requires N = 9-21 per cent.I-'henyltridecylnitrosoamilze is readily soluble in chloroform, ether,light petroleum, or benzene in the cold, sparingly 40 in cold alcohol,N = 9.41ENFIELD: THE REDUCTION OF CHLORIC ACID. 2441and crystallises from methyl alcohol in f awn-coloured, glisteningplates, melting at 3 9 4 0 O . It gives a well-marked Liebermann’sreaction, the colour changes being similar to those given by theother two nitrosoamines.CHEMICAL LABORATORY,ST. THOMAS’S HOSPITAL,LONDON, S.E