STEREOCHEMISTRY OF UNSATURATED CARBON COMPOUNDS. 81 V. -StereocImwis~~ y o f Uimturat ed Carbo 12 Compounds. Part I. EtheriJicntion of Substituted Acrylic Acids. By JOHN J. HUDBOROUGH and LORENZO L. LLOYD. IN a. series of communications presented to the German Chemical Society (Bey., 1894,27,510, 1580, and 3146), Victor Meyer and one of us were enabled t o show that, as regards the ease with which they are etherified, diortho-substituted benzoic acids differ completely from COOH their isomerides. Acids of the type x/\x , where X = C1, Br, NO,, COOH, &c., are entirely unacted on when their methyl alcoholic solu- tions are saturated with dry hydrogen chloride, either in the cold or a t the boiling point of the alcohol. Acids in which X = CH,, OH, F, &c., that is in which the weights of the substituting groups are small, yield minute quantities of their methylic salts when their boiling solutions are treated with hydrogen chloride for some time (V.Meyer, Rer., 1895, 28, 1259). V. Bleyer has since shown that the method of etheri- fication suggested b j 2, Fischer and Speier (Bey., 1895, 28, 3252), in I I VOL. LXXIII. u82 SUDROROUGH AND LLOYD : STEREOCHEMISTRY OF which the acid is boiled with a 3 per cent. solution of hydrogen chloride, yields similar results, and is preferable to the method first adopted, as many acids, which are but sparingly soluble or practically insoluble in cold methylic alcohol, dissolve readily on warming. It was stated by Victor Meyer and one of us that this abnormal behaviour of diortho- substituted benzoic acids is to be attributed to stereochemical causes, and Wegscheider (Monatsh., 1895, 16, 75) has since suggested that, if etherification be preceded by the formation of an additive compound of the acid and alcohol, the groups or atoms in the ortho-positions may be in such proximity to the csrboxylic group as to hinder or completely prevent the formation of the additive compound. This view is sup- ported by the fact that the radicle weight or volume plays an important part in the retardation or prevention of etherification (Meyer, Bey., 1895, 28, 1259).Recent investigations by Kellas (Zed. phys. Chenz., 1897, 24, 221) prove that the retardation is not merely due to the weight of the group in the ortho-position, as among ortho-substituted benzoic acids the nitro-group (NO, = 46) has a greater retarding influence than either bromine or iodine (Br = SO, I = 127).Menschutkin has also been able to draw generalisntions regarding the etherification of saturated acids of the aliphatic series from his researches on primary, secondary, and tertiary fatty acids * (Annalen, 1879,195,334, and 1879, 197, 193). Formic acid ...... 61.7 p. cent. Isobutyric acid ............ 29.0 p. cent. Acetic acid ...... 44.4 ,, Methylacetic acid ......... 21.5 ,, Propionic acid.. . 41.2 ,, Trimethylacetic acid ...... 8.3 ,, Butyric acid ... 33.3 ,, Dimethylethylacetic acid 3.5 ,, R \ From these results, it is evident that an acid with the grouping R,TC*COOH is much more difficult to convert into its ethereal salt than acids of the types RCH,* COOH and RR,CH* COOH, where R, R,, and R,, represent alkyl groups.The broader generalisation, that %I' acids of the type XI x \ -C COOH, where X not only represents alkyl XI/ groups but also C1, Br, NO,, &c., are difficult to etherify, does not, how- ever, hold good, since trichloracetic acid is more readily etherified than acetic acid itself. A reason for this anomalous behaviour of tri- halogenised acetic acid has been put forward by Feilmann and one of us (Proc., 1897, 241). * These numbers represent the initial velocity or the amount of acid (in percentage of the quantity originally present) converted into ethereal salt when nloIecular quantities of the acid and isobutylic alcohol are heated at 155" for 1 hour.UNSATURATED CARBON COMPOUNDS.PART I. 83 Although generalisations of the above nature have been made with reference to substituted benzoic acids, and also t o the fatty acids, no systematic study of unsaturated acids appears to have been made ; we therefore determined to prepare a number of the latter and to study the amounts of ethereal salt formed under different conditions, in order to find whether any general rules could be drawn from the results. A further incentive to this study was the suggestion made by one of us (Chem. News, 1895, 72, lS7) with reference t o the constitution of cam- phoric acid. I n that note, attention was drawn to the fact that Bredt's constitutional formula accounts for the characteristic behaviour of cam- phoric acid on etheritication with ethylic alcohol and hydrogen chloride, if the assumption be made that an acid with the grouping At the time this suggestion was put forward, no facts were known which justified the assumption, and it was partly with the object of determining whether further investiga- tions might supply satisfactory evidence on this point that the present research was started.Acids of the type mentioned are by no means oommon, whereas many acids are known of somewhat similar constitu- -7H2 7x0 COOH is difficult to etherify. -cx2 CXY CZ* COOH' tion, namely, unsaturated acids of the type I I We selected the latter class of acids as being the most suitable, and also because it seemed interesting to determine whether there was any great difference in the amounts of ethereal salt formed by the stereo- isomeric acids X* *H X- ;cI! *€I COOH* C *Y and Y*C*COOH' During the course of this investigation, a communication by Anschutz (Ber., 1897, 30, 2652) appeared whic' - ears on the same subject, with the exception that he investigated - -"w dicarboxylic acids, whereas we have restricted ourselves t o the study of monobasic acids.I n the summary a t the end of this paper, we discuss Anschiitz's results and compare them with our own. The following is a list of the acids we have investigated. Cinnamic acid, allocinnamic acid, atropic acid, ortho-, meta-, and para- a-Brornocinnamic acid and a-bromallocinnamic acid. The two isomeric P-bromocinnamic acids. The two isomeric up-dibromocinnamic acids. Dichloro- and di-iodo-cinnamic acids. nitrocinnsmic acids.84 SUDBOROUGH AND LLOYD : STEREOCHEMISTRY OX a-Cyanocinnamic acid, orthonitro-a-cyano- and metanitro-a-cyano- a-Phenylcinnamic acid and a-phenylallocinnamic acid.a-Phenylorthonitrocinnamic acid, a-phenylorthonitro-allocinnamic acid and the corresponding meta- and para-compounds. Triphenylacrylic acid. aj3-Di-iodoacrylic acid. The results obtained are given in tabular form a t pp. 91-92. I n some of our earlier experiments, we attempted to etherify the acids by saturating their methyl alcoholic solutions with hydrogen chloride in the cold and allowing the mixture t o stand for some time, but we found that this method was not adapted t o our purpose, as certain acids, especially nitrated acids, were almost insoluble in cold methylic alcohol and, therefore, yielded little 01- no ethereal salt by this treatment.I n all the later experiments, namely, those described in this paper, we used the Fischer-Speier met hod. A considerable amount of a 3 per cent. solution of hydrogen chloride in pure methylic alcohol (3 grams HCl in 100 grams of solution) mas prepared, and half a gram of the acid was boiled with 10 C.C. of this solution for an hour on the water bath, in a small flask fitted with a reflux condenser. For this purpose, the condenser described by Feilrnann and one of us (J. Xoc. Chem. Ind., 1897, IS, 979) is admirably adapted, as it does away with the possibility of moisture permeating the cork and thus vitiating the result. At the end of the specified time, water was added and the whole extracted twice with ether, any unaltered acid was removed by the aid of dilute sodium carbonate, and the ethereal solution, after drying with calcium chloride, was slowly distilled from a tared flask.The amount of ethereal salt formed was weighed after the flask had been standing over sulphuric acid for several hours. In all cases where the ethereal salts were solid, the residue was crystallised from methylic alcohol and the melting point taken. Oily ethereal salts were hydrolgsed, and the melting points of the acids thus obtained were determined. The melting points of recovered acids, in cases where etherification did not take place, or took place to but a slight extent, were also taken. cinnamic acid. Cinnanzic Acid, Allocinnamic Acid, and Atropic Acid. Ph.9.H H*c*Ph H*E-H H*C*COOH H*C*COOH Ph- C* COOH Cinnamic acid.Allocinnamic acid. Atropic acid. These three isomeric phlenylacrylic acids were first investigated. The The results, cinnamic acid obtained from Kahlbaum melted a t 133'. Nos. 1-8, obtained are given in the Table (p. 91).UNSATURATED CARBON COMPOUNDS. PART I. 85 The ethereal salt, after recrystallisation from methy lie alcohol, melted at 34" (Anschutz and Kinnicutt, Ber., 1879, 11, 1220, give 33.4'), and the regenerated acid obtained by hydrolysis melted at 133' (Kraut, Annalen, 1865, 133, 193, gives 133'). Allocinnamic Acid.-Prof. C. Liebermann, of Charlottenberg, was kind enough to provide us with about 2 grams of this acid, and we desire to express our thanks to him for his kindness. The acid melted at 68' (Liebermann, 68'). The results, Nos.9-14, are given in the Table, p. 91. The oily ethereal salt, after hydrolysis with warm potash, yielded an acid melting at 66-67'. It is thus evident that Fischer's method of etherification yields the salt of allocinnamic acid and not of cinnamic acid. Atropic Acid.-This acid, which we obtained from Schucharat, melted a t 106". The oily product, after hydrolysis with potash, gave an acid melting at 105'. The results, No. 15-22, are given in the Table. Ortho-, Afeta-, and Para-nitrocinnanaic Acids from Kahlbaum. Orthonitrocinrmnic Acid-See Table, Nos. 23 and 24. The ethereal salt, after crystallisation from methylic alcohol, melted a t 73' (Beilstein and Kuhlberg, Annalen, 1872, 163,126, give 72-73'), Metanitrocinnanaic Acid.-See Table, Nos. 25 and 26. This ethereal salt does not appear to have been described before.It crystallises from methylic alcohol in pale yellow prisms melting a t 123-124O. It is only sparingly soluble in methylic OF ethylic alcohol, and in ether or carbon bisulphide, but dissolves readily in chloroform or benzene. 0.500 gave 28.8 C.C. of moist nitrogen at 17" and 757 mm. N = 6.65. Theory requires 6.76 per cent. Paranitrocinnamic Acid.-See Table, Nos. 27 and 28. As the ethereal salt is almost insolublein ether, the amount formed could not be determined by the usual method. The process we adopted was as follows: water was added a t the end of the hour, the pre- cipitate collected, treated with dilute sodium carbonate solution in order to remove any unaltered acid, and then washed, dried a t loo', and weighed. After recrystallisation from alcohol, in which it is only slightly soluble, it melted a t 160' (Kopp, Jahyesbeyicht, 1861,410, gives 161").86 SUDBOROUGH AND LLOYD : STEREOCHEMISTRY OF a- am? p-Bromocinrzarnic Acids, Ph:R*H H*g*Ph Ph*fl*Br Br 9 -Ph Br*C*COOH Br-CGOOH H*C*COOH H*C*COOH a-Bromocinnamic a-Bromallo- B-Bromocinnamic B-Bromallo- acid.cinnamic acid. acid. cinnamic acid. The two a-acids were obtained by the method described by Scock- meyer (Diss., 1883), namely, by the action of alcoholic potash on cinnamic acid di bromide, and were separated by fractionally precipi- tating the solutions of their potassium salts with hydrochloric acid. The a-bromocinnamic acid, purified by crys tallisation of its sparingly soluble ammonium salt, melted a t 131O.Br = 35-21, C,H,*CH:CBr*COOH requires Br = 35.24 per cent. 0.200 gave 0.1656 AgBr. The results obtained with a-bromocinnamic acids, Nos. 29 and 30, The a-bromallocinnamic acid, after purification by recrystallisation 0.200 gave 0.1662 AgBr. Br = 35.36. Theory requires 35.24 per cent. The results obtained with a-bromnllocinnamic acid, Nos. 31-34, are given in the Table, As we thought the somewhat high numbers in 31-32 might be due to the presence of a small quantity of the isomeric acid melting at 131°, we took the acid recovered from the above experiments, and determined the amount of ethereal salt formed from this ; the results, Nos. 33 and 34, agree with those given above, The oily ethereal salts, when hydrolysed with cold aqueous potash, yielded an acid melting at 119O. From the fact that it crystallised from water in plates, that it dissolved with the greatest readiness in ammonia, and also from its melting point, the acid thus obtained was undoubtedly the a-allo-acid.From this, it is apparent that a-bromallocinnamic acid is converted into its true ethereal salt, and not into the isomeric a-bromocinnnmate, when etherified by Fischer’s method ; whereas, when its alcholic solution is saturated with hydrogen chloride, molecular rearrangement occurs, and the salt of a-bromocinnamic acid is obtained. The isomeric P-brom-acids were obtained by the method described by Michael and Brown (Ber., 1886, 19, 1379, and 1887, 20, 552), and we are able t o confirm their work in every respect; the two acids were separated by crystallieation, first from alcohol and then from chloroform, as these chemists recommend.are given in the Table (p. 91). from water, was obtained in glistening plates melting a t 1 2 0 O .UNSATURATED CARBON COMPOUNDS. PART I. 87 The P-bromocinnamic acid melting a t 134-135" crystalliscs from its chloroform solution on spontaneous evaporation in well-developed cubical crystals, 0.200 gave 0,1645 AgBr. Br = 35 0. Theory requires 35.24 per cent. The results obtained with this acid, Nos. 35-42, are given in the Table, p. 91. The oily ethereal salt, on hydrolysis, gave an acid which melted at 133-134", after recrystallisation from carbon bisulphide. P-Bronu;cZZocinnamic acid (see Erlenmeyer, Ansaalen, 1895, 287, l), melting a t 159O, is readily obtained pure after one recrystallieation from a small quantity of alcohol.0*2044gave0*1701 AgBr. Br = 35.42. Theoryrequires35.24 per cent, The results, Nos. 43-48, are given in the Table. The methylic salt crystallises from alcohol in thick, colourless prisms melting a t 58". It is moderately soluble in ethylic alcohol or benzene, and dissolves with great readiness in ether, chloroform or carbon bisulphide. 0.2423 gave 0.1887 AgBr. Br = 33.06. 0.2582 ,, 0.2019 AgBr. Br=33*28. Theory requires 33.19 per cent. The ease with which the two P-brom-acids can be obtained pure by following Michael and Brown's directions renders inexplicable such statements as those of Liebermann and Scholz (Ber., 1892, 25, 552) and of Erlenmeyer (ibid., 1886, 19, 1936) that the acid melting at 133-134' does not exist.Dichloro-, Dibromo-, alzd Di-iodo-cinnamic Acids, Ph*E'C1.-This acid Ph*g*Cl or C1 C* COOH COOH-C-Cl Dichlorocinnumic Acid, was prepared by saturating a chloroform solution of phenylpropiolic acid with chlorine. After recrystallisation from light petroleum, il melted a t 120' (Nissen, Ber., 1892, 25, 2665, gives 120-121"). The results, Nos. 49 and 50, are given in the Table. The ethereal salt was obtained in the form of an oil. Bib~*omocinnarnic Acids.-Two isomeric dibrom-acids were obtained by adding bromine to phenylpropiolic acid in chloroform solution (Roser and Haselhoff, Annalew, 188S, 24'7, 139) ; they were separated by dissolving them in a small quantity of chloroform, and adding lightS8 SUDBOltOUGH AND LLOYD : STEREOCHEMISTRY OF petroleum until a permanent turbidity was produced. The acid melting a t 139" crystallised first, and was purified by recrystallisation from boiling light petroleum (b.p. 60-80'). The isomeric acid melting a t 100' ma8 obtained pure only after repeated solution in chlorofcrm and precipitation by light petroleum. The results with the acid melting at 139" (Nos. 51 and 52), and with the acid melting a t 100' (Nos. 53 and 54), are given in the Table. The acid recovered from the former melted at 139". Di-iodocinnamic Acid, CIPh:CI* COOH, obtained by the method given by Liebermann and Sachse (Ber. 1891, 24, 4113), after several recrvtallisations from chloroform, melted at 167'. (1,. and S. give 171"). The results, Nos. 55 and 5 6 , are given in the Table (p. 92).The recovered acid melted at 170". a-Cyanocinnamic Acids. a-Cyanocinnamic acid, Ph0;CI.H or obtained NC*C*COOH CNCCOOH' by Carrick's method (J. pr. Chem., 1892, 45, 401), after recrys- tallisation from alcohol, melted at 180' (Carrick, 180'). See Table, Nos. 57 and 58. The methylic salt thus obtained crystallised from its alcoholic solution i n small, colourless prisms melting a t 89'. It is readily soluble in chloroform or ether, and moderately in alcohol, benzene, and carbon bisulphide. 0.5 gave 31.8 C.C. moist nitrogen at 13' and 752 mm. CHPh:C(CN)*COOMe requires 7.48 per cent. a-Cyctlzo-orthonitroc~nnccm~c Acid-The ethylic salt of this acid was obtained by the action of sodium ethoxide on a mixture of ethylic cyan- acetate and orthonitrobenzaldehyde (Riedel, J.pr. Chem., 1896, 54, 541), and was hydrolysed by the requisite quantity of normal sodium hydroxide a t about 60". The acid thus obtained melted at 223' (Riedel, 223'). The results obtained are given in Nos. 59 and 60 of the Table. The recovered acid melted at 220". The methylic salt crystallised from alcohol in small, discoloured needles melting a t 142'. a-Cyanometanitrocinnamic acid was obtained in a similar manner. We find that the ethylic salt melts a t 134', and that the melting point is not altered by repeated crystallisation from alcohol. (Riedel gives 0.4 gave 39.4 C.C. moist nitrogen a t 16" and 761 mm. N = 7.43. 1 2 7-1 2s'). N = l l 5 . N02*C,H,*CH:C(CN)*COOEt requires N = 11.38 per cent.TJNSATURATED CAREON COMPOUNDS. PART I , 89 The resulh, Nos.61 and 62, obtained with this acid are given i n the Table. The recovered acid melted a t 17O-17lo(Riedel, 172'). The methylic salt, after recrystallisation from alcohol, was obtained in fine, silky needles melting at 135-1 36", and readily soluble in the usual solvents. 0.3 gave 30.8 C.C. moist nitrogen a t 16' and 762 mm. N = 12.0. NO,*C,H,*CH:C'(CN).COOMe requires N = 12-07 per cent. a-Phen&innamic Acids, Ph*Q*H H*y*Ph Ph*C*COOH Ph*C-COOH a Phenylcinnamic acid. a-Phenylallocinnnniic acid. These acids were prepared and separated by Bskunin's method (Gaxx., The a-phenylcinnamic acid, melting a t 173", gave the results Nos. 63 The ethereal salt, after recrystallisation from alcohol, melted a t 77" The a-phenylallocinnamic acid, after crystallisation from water, was 1897, 27, ii, 48).and 64 of the Table. (Bakunin, 77"). obtained in colourless prisms melting a t 136'. The yield of allo-acid was extremely small. 0.5 gram, after 1 hour with 3 per cent. solution, gave 0.2137 gram ethereal salt. The methylic salt thus obtained was semi-solid, whereas Bakunin, who obtained the same compound by the action of methylic iodide on the silver salt of the acid, describes it as an oil. The solid we obtained apparently contained a considerable amount of the solid ethereal salt of the isomeric a-phenylcinnamic acid, as tho acid recovered from this first etherification yielded an oil when treated a second time with the 3 per cent. hydrogen chloride solution. The results are given in Nos. 65 and 66 of the Table. a-Phenylorthonitrocinnamic Acids.-A mixture of the two isomeric acids was obtained by the action of acetic anhydride and sodium phenylacetate on orthonitrobenzaldehyde (Bakunin, Gaxx., 1895,25, i, The recovered acid melted a t 137-138'.NO,*C!,H,*fi*H Ph*C*COOH' was 137). The a-phenylorthonitrocinnamic acid, readily obtained pure by Bakunin's method. It melted a t 195-196', and on etherification gave the results Nos. 67 and 68 of the TabIe. The ethereal salt, after recrystallisation from alcohol, melted a t 7 5 O (Bakunin, 75-76'),90 SUDBOROUGH AND LLOYD : STEREOCHEMISTRY OF A mixture of the above acid with the isomeric allo-acid was obtained by Bakunin's method; it melted a t 155", whereas the pure allo-acid melts at 146-147'. We adopted the following method for its purifica- tion.The mixture of acids was boiled for 1 hour with a 5 per cent. solution of hydrogen chloride in methylic alcohol, 20 C.C. of this solution being used for each gram of the mixed acids; after etherification, the solution was poured into water, extracted with ether, and the ethereal solution washed with sodium carbonate solution ; on the addition of hydrochloric acid to this alkaline solution, the a-phenylorthonitroallo- NO,*C,H,*Y*II cinnamic acid, , was thrown down, and after recrystal- COOH*C*Ph lisation from dilute acetic acid melted a t 146'. Whether the results, Nos, 69 and 70 in the Table, are somewhat too high owing to the presence of a small quantity of the isomeric acid, we cannot at present say with any degree of certainty, as the yield of allo- acid is extremely poor and we had but a gram or so at our disposal.a-Phenylrnetnnitrocinnanaic Acids.-The two acids were obtained by Balrunin's method (Zoc. cit.). The a-phenylmetanitrocinnamic acid melted a t 181' and gave the results Nos. 71 and 72 in the Table (p. 92). The ethereal salt, after recrystallisation from alcohol, melted a t '72'. a-PhenyZnaetnnitroaZlocinnarnic acid melted a t 196". The results, Nos. 73 and 74, obtained with this acid are given in the Table. a-P~~enylIoarc~nitrocin?zccr~zic acid melting a t 2 14' gave the results Nos. 75 and 76 of the Table. The ethereal salt crystallised from alcohol in yellow needles melting a t 141-142' (Bakunin, 141-142'). a-Phenyi'paranitronllocinnanzic acid melted a t 144' (Ba,kunin, 142'). For results, see Nos.77 and 78 of Table. The recovered acid meltedat 144'. Triphenylacrylic acid, CPh,:CPh*COOH, was prepared from the acid amide (Heyl and Meyer, Bey., 1896,29,2786) by a method very similar to that adopted by Heyl and Meyer, except that we used the amide, in a fine state of division, suspended in dilute sulphuric acid, and kept the mixture well stirred by an automatic stirrer while the nitrite solution was being run in ; the acid thus obtained melted a t 213'. The results of etherification, Nos. 79 and 80, are given in the Table. The unaltered acid melted a t 213-214". Heyl and Aleyer (Zoc. cit.) have already shown that this acid is only slowly etherified when hydrogen chloride is passed for several hours through its solution in boiling methyl alcohol. ap-Di-iodacqZic acid, CHI :CI*COOH.-This acid was prepared by the addition of iodine to propiolic acid, as described by Bruck (Ber., 1891, 24, 4120) ; after recrystallisation from chloroform, it melted a t about 7 6 O , but after recrystallisation from water at 104' (Bruck, 104').1 2 3 4 5 6 7 a 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 29 30 31 32 33 34 35 36 37 38 39 40 4 1 42 43 44 45 46 47 48 28 10 UNSATURATED CARBON COMPOUNDS.Table of Results. o * A l < A PART I. 0.5 0.5 0 -5 0.5 0.5 0.5 0.5 0 *5 0 -5 91 Name of acid. Cinnaniic. Do. Do. Do. Do. Do. Do. Do. Allocinnamic. Do. Do. Do. Do. Do. Atropic. Do. Do. Do. Do. Do. D O . Do. Orthonitrocinnnmic. Do. Metanitrocinnamic. Do. Paranitrocinnamic. DO. a-Bromocinnamic. Do. a-Bromallocinnamic. Do. Do.Do. 8-Bromocinnamic, m. p. 134-135". Do. DO. Do. Do. Do. Do. Do. 3-Bromallocinnamic, m. p. 159". Do. Do. Do. Do. Do. ;ram. 0.5 0 -5 0.5 0 *5 0.5 0.5 0.5 0.5 0.5 0 *5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0-5 0-5 0 -5 0-5 per cent. 3 3 3 3 1 1 1 1 3 3 3 3 1 1 3 3 3 3 1 1 1 1 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 1 1 1 1 3 3 1 1 1 1 I 10 ' 1 0 ' 10 , 10 j :; I ;: 10 1 0 10 10 1 0 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 1 0 10 1 10 - z .* i. .iiiiis. 60 63 10 10 60 60 30 30 60 60 10 10 60 3 0 60 60 10 10 60 60 30 30 60 60 60 60 60 60 60 ti0 60 60 60 60 60 60 60 60 10 1 60 10 70 10 10 10 10 10 10 60 30 30 6C 60 60 60 30 gr. 0.5425 0.5417 0.4958 0'4982 0.5433 0'5459 0.4996 0.5033 0'5282 0'5304 0'4269 0'4261 0'4820 0.4692 0.5102 0.5112 0.3420 0.3421 0.4442 0'4408 0,2989 0.3023 0.3176 03195 0.5227 0,5203 0.5223 0.5218 0'4091 0'4038 0.1312 0'1302 0.1332 0.1312 0.5291 0.5308 0.5256 0'5281 0.4666 0'4698 0'4222 0'4231 0.5241 Os5260 0'4648 0.4630 0.4165 , "" , " L A V A I A" I , I 1 I 2 2 2.zi 0% 4 x 3 g g 5 sz W ' c d 99.12 99.64 90.60 91-02 99-28 99.74 91.25 91 '96 96.52 96 -92 18.00 77.86 88.06 85 '73 98.22 93 '40 62.50 62.52 81.16 80.54 54.62 55.23 59.22 59.58 97'48 97.02 97.40 97.30 77.06 76-06 24-72 24.52 25-09 24'72 99.68 100*00 99.02 99.48 87.90 88.50 79.54 79.70 98.74 99.08 87.56 87-22 78.46 78-2692 XTJDBOROUGH AND LLOYD : STEREOCHEMISTRY OF TabZe OJ Results-(continued).49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 - Name of acid. Dichlorocinnaiiiic.Do. Dibromocinnamic, in. p. 139". Do. Iibromocinnamic, m. p. 100" Do. Di-iodocinnamic. Do. a- Cyanocinnamic. Do. a-Cyano-orthonitro- cinnamic. Do. x-Cyanometanitrocinnan~ic. Do. a-Phenylcinnaniic. Do. a-Phenylnlloeinnamic. Do. a-Phenylorthonitro- cinnamic. Do. a-Phenylorthonitroallo- cinnamic. a-Phenvlorthonitroallo- cinnamic. .-Phenylmetanitrocinnamic Do. a- Phenylmetanitroallo- ciunainic. Do. a-Phen ylparani troci 11 nmnic Do. a-Phen ylparanitroaI10- cinnamic. Do. Triphen ylacry lic. Do. aj3-Di-iodacry lic. Do. gram. 0'5 0.5 0 -5 0.5 0.5 0 '5 0.5 0.5 0.5 0-5 0.5 0.5 0.5 0.5 0.5 0.5 0-5 0.5 0.5 0 -5 0.5 0.5 0 *5 0 -5 0.5 0.5 0 ' 5 0'5 0.5 0 -5 0 '5 0.5 0.5 0.5 - per cent. 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 C.C.10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 - mins. 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 - gr . 0.1645 0'1615 0'0414 0.0428 0.0700 0,0687 0'0109 0.0102 0.3466 0'3472 0.2677 0,2702 0.3203 0.3176 0-3981 0.4013 0.0631 0.0596 0.3974 0,3982 0.0813 0.0795 0.3829 0.3849 0'0744 0.0710 0,3746 0.3804 0-0684 0.0657 0'0112 0'0124 0.0387 0.0352 - 30 -90 30'34 7 -92 8-19 13.38 13'13 2.10 1-96 64'14 64'24 50'32 50.78 60'20 59.70 74-94 75.54 11'88 11'22 75.55 75.70 15'45 15'11 72.79 73'17 14.14 13'53 71-21 72.32 12'99 12.48 2'14 2 3 7 7'42 6-75 The results obtained with this acid, Nos. 81 and 82, are given in the Table. The recovered acid melted at 103-104".UNSATURATED CARBON COMPOUNDS.PART I. 93 Discussion of Results. From the results of the experiments described in the pi*eceding pages, we consider ourselves justified in drawing the following con- clusions. 1. Unsaturated acids which, in addition to an a-substituting group, also contain a radicle in the cis-position relatively to the carboxylic group, that is, acids of the type H*G*X Y C* COOH are difficult to etherify when boiled for an hour with a 3 per cent. solution of hydrogen chloride in methylic alcohol. The same property is also characteristic of acids in which the third hydrogen atom of acrylic acid is replaced by a substituting group z* E'X Y* C-COOH' 2. Di-substituted acrylic acids in which one group is in the a-position and the other in the t~ans-position relatively to the carboxylic group, are readily etherified under the conditions given above.Y*I;;'*H X* C*COOH This remarkable difference in behaviour supplies us with a simple method of determining the configurations of stereoisomeric acids, CHX:CY*COOH, where X and Y may be alike or dissimilar. We require merely to boil half a gram of each acid with 10 C.C. of a 3 per cent. solution of hydrogen chloride in methylic acid, and determine which acid yields the larger percentage of ethereal salt. This acid will be the one with the substituting group in the trans-position, and the acid which yields little or no ethereal salt wiil have the substituting group in the cis-position. We also think it probable that this difference on etherification may be made use of in the separation of such stereoisomeric acids, in very much the same manner as Martz (Ber., 1894, 27, 3147) and Jannasch and Weiler (ibid., 3447) have been able to separate diortho-substituted benzoic acids from their isomerides.The separaT tion in the acrylic series will not be so complete as in the benzoic, as, according to our results, the difference on etherification is not so marked as in the benzoic series. We have found that the method can be used with advantage in the separation of a-phenylallocinnamic acid from a-phenylcinnamic acid, and it can undoubtedly be used with equal94 SUDBOROUGH AND LLOYD : STEREOCHEMISTRY OF advantage in the separation of the corresponding nitro-acids. The method adopted by Bakunin (Zoc. cit.) for the purification of these acids is lengthy and tedious, and can probably be considerably curtailed by the process of etherification.We may point out that our results confirm in a remarkable manner the constitutions of the a-phenylcinnamic acids arrived a t by Bakunin from entirely different considerations. 3. Substituted acrylic acids in which the substituting groups are only in the P-position are readily etherified under the conditions given above. As examples of this generalisation, we have the two P-bromo- cinnamic acids, both of which yield over 90 per cent. of ethereal salt. We are a t present engaged in preparing PP-di-iodacrylic acid, and hope shortly to be able to state that this obeys the same lam. 4. The results we have obtained with mono-substituted acrylic acids are somewhat too meagre for us to draw general conclusions with any degree of certainty ; those, however, which we have so far obtained by using more dilute solutions of hydrogen chloride, namely, a 1 per cent., in methylic alcohol, seem to indicate that a n a-substituted acrylic acid is more difficult to etherify than the isomeric P-compound.I n support of this, we have the fact that atropic acid (a-phenylacrylic acid) is more difficult to etherify than either of the P-phenylacrylic acids (cinnamic and allocinnamic acid), This conclusion is further supported by Anschutz’s results (Ber., lS97’,30,2652). Anschiitz finds that mesaconic acid, Me*f?cooH, when boiled for a short time COOH-C-H with a 0.5 per cent. solution of hydrogen chloride, yields the mono- Me* R*COOH methylic salt, C0OMe.C-H The differences between cinnamic and allocinnamic acids point to the conclusion that a P-substituted acrylic acid, in which the sub- stituting radicle is in the cis-position relatively to the carboxylic group, is more difficult to et,herify than the isomeric trans-compound.We give these generalisations with the greatest reserve, as further investigations with other acids, for example, crotonic acids, are necessary to prove whether they are correct. 5. The radicle weights or volumes of the substituting groups in the a- and cis-positions appear to be an importaat factor in determining the actual amount of ethereal salt formed in each case. A survey of the results obtained with dichloro-, dibromo-, and di-iodo-cinnamic acids brings out this generalisation with great clearness. The dichlor- acid yields more ethereal salt than either of the two dibrom-acids, and these again yield more than the di-iodo-acid.This conclusion is entirely in accordance with V. Meyer’s work on diortho-substituted benzoic acids, and with that of Kellas (2. physik. CThern, 1897, 24,UNSATURATED CARBOK COMPOUNDS. PART I. 95 221) on mono-substituted benzoic acids, and also with our own on substituted benzamides (Trans., 1895, 233). 6. The presence of a nitro-group in the ortho-position in cert'ain cinnamic acids, for example, in orthonitrocinnamic acid itself, and also in a-cyano-orthonitrocinnamic acid, appears t o have a retarding influence on the formation of the ethereal salt. This is in complete harmony with a suggestion made by Victor Meyer and one of u s several years ago, but which received no support from actual experi- ment conducted a t that time (Bey., 1895, 28, 1267).It is a well known fact that ortho-substituted cinnamic and hydrocinnamic acids readily undergo condensation, yielding ring compounds. For example, CH /\/\CH \/\//C*oH /\GH:CH* COOH gives I I I N CH', The fact that the isomeric meta- and para-compounds undergo no similar condensations is supposed t o be due to the fact that, in these acids, the substituting groups are not sufficiently near to the car- boxylic group t o allow of the elimination of H,O, HBr, &c The results we have obtained may be due to the fact that the nitro-group is in closer proximity to the carboxylic group in the ortho-acids than in the meta- and para-acids, and this may account for the retardation.If so, we should expect to meet the same phenomenon in all ortho- substituted cinnamic acids, and also in diortho-substituted cinnamic acids. This is a point which we consider deserves a little more atten- tion, and we purpose studying a number of these acids. 7. The results we have obtained by the etherification of allocinnamic acid and a-bromallocinnamic acid indicate that Fischer's method of etherification yields the ethereal acids of the allo-acids, and not those of the more stable isomeric acids. This is an extremely interesting point, since other authorities state that these allo-acids, when their alcoholic solutions are saturated with hydrogen chloride and allowed to stand, yield the ethereal salts of the more stable acids.8. In the introduction to this paper, we stated that one of the reasons for undertaking the investigation was t o account for the characteristic behavionr of camphoric acid on etherification by the aid of the stereochemistry of the acid molecule. As the result of our96 STEREOCHEMISTRY OF UNSATURATED CARBON COMPOUNDS. investigation, we are able t o state that in unsmturated acids a carb- oxylic group which bas substituting groups in the a- and cis-positions with respect to itself is difficult to etherify. It is true that in Bredt’s formula for camphoric acid, and also in the newer formula suggested by W. H. Perkin, jun. (Proc., 1897, ZlS), one of the carboxglic groups is thus situated, i t has substituting groups in the a- and also in the cis-position. The other carboxylic group in camphoric acid is not so situated; it has a substituting group in the cis-position, but none in the a-position. We consider, then, that our results render the behaviour of camphoric acid on etherification explicable if we adopt either Bredt’s or Perkin’s formula. The same remarks, however, do not apply to Tiemann’s formula (Bey., 1895, 28, 1079) : Me2- 7H2 (?Me, YH* COOH UHMe- CH* COOH CH,* YH-COOH CH,* CMe*COOH CMe(CO0H)-CH, The great difference is that camphoric acid is a ring compound, whereas our researches have been limited to aliphatic unsaturated acids. 9. The results we have obtained are in complete harmony with the configurations of unsaturated acids according to the van’t Hoff - Wislicenus theory, and we consider that they establish with certainty the conclusion previously arrived at, namely, that in what are usually termed cis-substituted monocarboxylic acids the substituting group is in closer proximity to the carboxylic group than when it is in the trams-posi t ion. I n conclusion, we may state that, having obtained such interesting results with monocarboxylic acids, we a t once turned our attention to dibasic acids. From a private communication from Professor Anschiitz, we learn that he has already taken up the study of a number of such acids in the direction indicated in the Berichte, and we have therefore not investigated any of these acids ourselves. The question whether generalisations similar to those we have obtained for unsaturated acids may not also hold, to some extent, for saturated acids immediately presented itself to us, and the fact that such acids as dibromosuccinic acid, dibromhydrocinnamic acid and it,s nitro-derivatives, are difficult to etherify indicates that interesting results may probably be obtained in this direction, FH*COOH 1 $JH2 Perkin. Tiemann , P 1 ?Me2 Bredt. UNIVERSITY COLLEGE, NOTTINGIIAM.