TIA PREPARATION AND STlLUCTUZLLZL "F A DILER CF PAI\BACIOOm ~32": 1,3—DI-(PARAEROIOEJI ANYL) -l-BUTEhE By John wallace'Pearce A TI‘SIS Submitted to the School of Graduate Studies of hichigan State College of Agriculture and Apglied Science in partial fiulIillment of the requirements for the degree of uASTdR OF SCIiK Degertment of Chemistry 1950 x «:34 | .7 ._-. The writer wishes to express his sincere gratefulness to Dr. Gordon L. Goerner whose tolerant and kind guidance made this work possible. 0 5-1” , | :y: I: ‘ I j a . .. ”a . ».:» I; =' ‘5". . . , . ‘\ TABLE or commrrs PAGE IIQTRODUCTCIOI‘IOOOCOOOO OOOOOOOOOOOOOOOOOOO0.00... l EJG‘ERIMLJNTAL................................... 7 Chemicals................................. 7 p-Bromophenylmethylcarbinol............... 9 quhloro-o(-(p-bromophenyl)-ethane........ 13 1,3-Li-(p-bromophenyl)-l-butene........... 16 Degradation Studies on l,3-Di(p- bI‘Othphenyl)-l-bUt€De. o o o o o o o o o o o o o o o o o o o o 21 PREPARATION OF REFEREICE ALDEIKDES............ 27 p-Bromobenzaldehyde....................... 28 p-Bromohydratropaldehyde.................. 30 DISCUSSION.................................... .7 SU4fiAflY....................................... PfiaFiliED-ICESO0.000000000000000DOOOOOOOOOOOOOOOOO 44 INTRODUCTION In an attempt to prepare hydratrogonitrile (« - phenylpropionitrile) I, Goerner and Hines (l) Cfis-CH-CN I heated q-chloroethylbenzene with cuprous cyanide to a temperature of 130°C. At this temperature a vigorous re- action ensued with the evolution of hydrogen cyanide. The product proved to be cis-l,3-diphenyl-l-butene, II. CH=CH ~CH-CH3 Q A search of the literature showed no previous report of II this reaction. The purpose of this investigation was to prepare (-chloro-a(-(p-bromoghenyl)-ethane, to attempt to react it with cuprous cyanide, and to determine the struc- ture of any reaction products obtained. A study of the paper mentioned above (1) together with the report submitted by Eines brings out the follow- ing points: I. The reaction between«(-chloroethylbenzene and cuprous cyanide proceeds quite vigorously at a tempera- ture of about 130°C. 2. Lowering the temperature of reflux by the use of a solvent with a boiling point below 1300C. inhibits the reaction. 8. An anhydrous medium is necessary. If water is present, the hydrocarbon is not obtained, but rather, di- c(-methylbenzyl ether C‘Hg-CH(CH3)-O-CH(CHJ)-C‘Pg, III, (2). 4. The yield of gig-l,E-diphenyl-l-butene is about 40%. 5. The composition of the copper-containing resi- due has not been investigated. Later work (2) has shown that the reaction between the Q-chloroethylbenzene and cuprous cyanide will proceed in a solvent if the boiling point of the latter is sufficiently high. Commercial xylene and styrene have been used as sol- vents. The use of styrene has brought about an increase in the yield of the product. By using a large excess of the latter, more gig-l,3-diphenyl-l-butene was obtained than could be obtained from thed-chloroethylbenzene alone. Ob- viously some of the styrene used as a solvent was entering into the formation of the gig-l,S-diphenyl-l-butene. The structure of the hydrocarbon obtained by Goerner and Hines was easily proved because of the following facts. 1. The physical properties of the hydrocarbon approximated those for the already described dimeric Sty- renes or distyrenes (3). 2. Hydroxflation and degradation of the hydro- carbon gave benzaldehyde (characterized as its 2,4-dinitro- phenylhydrazcne). The position of the double bond in the hydrocarbon was thus determined, and on this basis it be- came apparent tnat formula IV or V represented the compound under consideration. cazca-ca-cns CH=CH-CH2-CH2 IV V No second degration product was isolated. a. A dibromide of IV, m.p. 102°, was described in the literature (4). 4. The compound under consideration formed a di- bromide melting at 10230. when alone or when mixed with an authentic sample of the already described dibromide. 5. Marion (5) had shown that the trans form of IV melted at 47°C., whereas the gig form was a liquid. Erlenmeyer (7) prepared 1,3-diphenyl-l-butene (IV) by the action of HBr and cinnamic acid and Swartz (7) did the same using hCl. Fittig and Erdman (6) refluxed cinnamic acid with various concentrations of sulphuric ac1d, extracted the reaction mixture with ether and recovered the distyrene (IV) from the extraction. Stobbe and Posnjak (9) repeated hrdman's work and proved the structure of the distyrene (IV) on the basis of the iso- lation of benzaldehyde and comparison of the ultra violet light absorption spectra data with Other ethylenic compounds. ‘4‘ U0 Stormer and Kootz (4) brominated IV without significant substitution in cold C52 solution to give 1,2-dibromo-l,5- diphenyl-butane (VI). Br Br HC - CH — CH-CHB O 9 Three of four possible isomers were separated by crysta- VI. llization from petroleum ether. Two of the isomers were shown on mixing to melt at 102°C” the reported melting point of the brominated gig-l,B-diphenyl-l-butene (IV). They further showed isomerization of 1,3-diphenyl-l-butene (IV) to 1,3-diphenyl-2-butene on long standing or heating with high concentrations of sulphuric acid. Kdnigs and hai (10) have prepared the distyrene (IV) by the action of glacial acetic acid and sulphuric acid on sty- rene. More recently 1,3-diphenyl-l-butene has been prepared by marion (0). he prepared (,Y—diphenylbutyric ac1d by a Clezmnensen reduCtio-n of (-phenyl- \Lbenzoylgropionic acid. A Bouveault reduction of the substituted butyric ester gave «,Y -diphenylbutyl alcohol. A potassium bisulphate dehydra- tion yielded l,3-diphenyl-a-butene which isomerized upon standing to l,udiphenyl-l-butene (IV). he alSU obServed that l,d-diphenyl-l-butene, upon Standing With sulphuric acid, isomerized to 1,3-diphenyl-2-butene. Marion's product, a solid, melting at 47‘C., was assigned the pgaps configuration. He further states that the gig and traps forms of 1,3-diphenyl-l-butene (IV) each have two possible dibromides, the four melting points being 79°, 86.5: l22°and 129°C. The last two are those formed by the gig-distyrene, which is a liquid. Further, a mixture of the two gig-dibromides melts at 102°C. This is in agreement with the results of Stoermer and Kootz (4). Barber et al. (ll) prepared (-phenyl-K-chloroethane and Condensed it to mgsp-2,3diphenyl-butane by means of magnesium in boiling ether. (—(p-Bromophenyl)-a(-chloro- ethane was prepared from the corresponding carbinol by treatment with thionyl chloride but would not react with magnesium. A Wurtz type condensation was carried out by refluxing this chloride with sodium wire in anhydrous benz- ene for sixty hours. figsp-2,8-di(p-bromophenyl)-butane was obtained. Barber et al. brominated megp-2,3-diphenylbutane in a glacial acetic acid-water mixture under reflux. A 25% yield of a racemic 2,B-dibromo-2,B-di-(p-bromophenyl)‘ butane was obtained. This tetrabromo compound was partially debrominated with some difficulty by suspending it in glacial acetic acid under reflux and adding zinc dust. Pre- dominatelx gammibmmo- «,fi-dimethylstilbene was ob- tained. The trans isomer was obtained by refluxing the gig isomer in nitrobenzene which contained a tr ce of iodine. The reactivity of the aromatic bromine was demon- strated by conversion of trans-4,4ldibromo-¢(,F;-di- methylstilbene to 4,4J-dicyano-fi , p -dimethylstilbene by treatment of the former with cuprous cyaiide in boiling quinoline. Seventy—four percent conversion to the cyano derivative was obtained. In Spite of the scarcity of information about dimers of p-bromostyrene, it was hOped to proceed with the investigation in a manner similar to that used by Goerner and Hines. Because of the discovery of several discrepancies in the literature, it became necessary to prepare two reference aldehydes and their derivatives. EXPERIMENTAL The following chemicals were used in this investigation: p-Dibromobenzene, Dow, m.p. £4 - 85. Diethyl ether, C.p. dried over sodium wire magnesium turnings, for Grignard reagents Ethyl bromide Hydrogen chloride gas, anhydrous from a cylinder Methyl alcohol, C.p. anhydrous Merck Benzene, thiophene free, C.p. anhydrous 2,4-Dinitrophenylhydrazine, Eastman white label Cuprous cyanide, J.T. Baker purified Tertiary butyl alcohol, Eastman white label Hydrogen peroxide, 30% C.p., Baker and Adamson Osmium tetroxide, Eastman white label p-Bromoacetophenone, Eastman white label Ethyl chloroacetate, Eastman white label Sodium metal, Merck, reagent Ammonia, commercial grade from cylinder p-Bromotoluene, Dow, commercial grade Bromine, Dow, N.F. Acetaldehyde- acetaldehyde was prepared just before use by depolymerizing paraldehyde. Toluene- sulfonic acid and concentrated sulfuric acid were used in small amounts as depolymerizing agents. Paraldehyde was heated with the de- polymerizing agents and acetaldehyde distilled slowly from a 72 cm. Vigreux column at 20 - ZIP. The receiver was chilled with crushed ice to prevent evaporation of acetaldehyde. p-BROAOPHENYLMETHYLCARBINQL Parabromophenylmethylcarbinol IX has been prepared previously by two different methods. Ziegler and Tiemann (12) and later Quelet (13) prepared the mono-Grignard re- agent of p-dibromobenzene and treated this with acetalde- hyde. Quelet reported yields of 45 to 50%. Overberger and co-workers (14) prepared the carbinol from p-bromo- benzaldehyde and methylmagnesium bromide in yields of 64%. Because of the availability of p-dibromobenzene, the former method of preparation was used throughout this work. The following equations illustrate this prepara- tion. dry BrOBr 4 Mg ———) Br MgBr ether VII H BrOm‘gBP + BBC-CHO-flBrO-C-CHg (SIVIgBP VIII H H7 H BrOC'J-CH31- H20 6 IBrQC'I-CH3 ongsr 0H IX In a typical preparation, 108 gms. (4 moles) of magnesium were placed in a 5-1. three-necked flask equipped with mechanical glass stirrer, reflux condenser and an externally cooled Hershberg dropping funnel. To the flask was added 150 ml. of anhydrous ether followed by 5 ml. of ethyl bromide to serve as the "starter". To this, a saturated solution of 944 gms. (4 moles) of p-di- bromobenzene in about 1500 ml. of anhydrous ether was added over a period of six hours. The upper surface of the flask was cooled from time to time with ice. After the Grignard reagent had been stirred for two hours, the flask was placed in an ice bath and a solution of 180 gms. of freshly distilled acetaldehyde in 700 ml. of anhydrous ether was added. The dropping funnel was chilled by an ice-salt mixture to prevent evaporation of the acetalde- hyde. The time of addition was 2% to 3 hours. Stirring was continued for one hour and the reaction mixture was then permitted to stand about thirty-six hours. During this time a dark green viscous layer separated from the ether. Hydrolysis was accomplished by pouring the reaction mass upon crushed ice. The ether layer was decanted and the aqueous layer acidified to Congo paper with dilute H2304. The aqueous phase was extracted with three 500 ml. portions of ether. These latter ether extracts were 10. washed with 10% sodium carbonate solution until alkaline and then added to the first ether decantate. The combined ether solution was dried over anhydrous potassium carbon- ate and the ether stripped from the solution. The carbinol fraction was distilled under reduced pressure through a 62 cm. Vigreux column. The following fractions were collected: . - r 0 Fraction at. Gms. B.P.,C. P., mm, nl8 ._____._ M_______ _______ _______ .y__ I - 25- 74 13 - II 28.3 75-106 2 1.5752 III 148.2 lO9-110.5 0.5 1.5728 IV 79.5 111-123 1.5 1.5780 V 180.0 residue - - Dehydration of the p-bromophenylmethylcarbinol IX could be minimized by distillation through a shorter column. A reasonably pure product was obtained by dis- tillation from a Claisen flask. The following table of data summarizes the results of the preparation of p-bromophenylmethylcarbinol from dibromobenzene. ll. REACTANTS YIELD l8 fig? 22::g§::§32 Gms.Mg. Gms. % DD gggidue 1. 944/180 108 451.6 56.4 1.5700 145 2. 708/135 81 227.7 37.8 1.5728 180 3. 708/135 81 115.7 17.5 1.5745 171.6 4. 785/167 90 278.0 41.7 1.5710 153.3 6. 944/180 108 247.7 30.8 1.5705 263.3 7. 944/180 108 368.2 45.8 1.5700a 90.0 a) temperature 20 0. Run number 7 was refluxed one hour after the dibromo- benzene had been added and the carbinol residue was dis- tilled from a Claisen flask. Numbers 6 and 7 were sampled before addition of ace- taldehyde and an aliquot was titrated for yield of p-bromo- phenylmagnesium bromide using standard acid—base titration. Yield of p-bromophenylmagnesium bromide in run no. 6 was 89.9%, in no. 7, 91.6%. 12. c(-CHLORo-o(—(p-BaomoPHnNYL)~ETHANE Barber et al. (11) have prepared<(-chloro-0(-(p- bromopheny1)-ethane X in 81% yield by heating p-bromo- phenylmethylcarbinol IX and thionyl chloride on a steam bath. They reported the product to be a pinkish pung- ent oil boiling at 115-12o°/ 11-12mm. In this work, p-bromOphenylmethylcarbinol IX was converted toaK-chloro-°(-(p-bromopheny1)-ethane X accord- ing to the following reaction: Br<::>+ca - CH3+ dry HCl gas-———+ Br<::;>0H - CH3+ H20 0H Cl IX X This reaction was always carried out in a well ventilated hood. A typical run is as follows: One hundred to three hundred grams of p-bromOphenylmethyl- carbinol IX was placed in a 500 m1. three-necked round bottom flask equipped with a mechanical stirrer, an in- let tube which was submerged below the surface of the p-bromophenylmethylcarbinol, an outlet tube attached to a calcium chloride drying tube, and a thermometer. The flask and its contents were chilled to 0°C. by means of a sodium chloride-ice bath. H01 gas from a cylinder was passed first through a trap, through a concentrated sulphuric acid wash bottle, through a second trap 13. and finally into the reaction flask at a slow, con- tinuous rate so that bubbling could be detected. The reaction temperature was not permitted to change appre- ciably from 0°C. Introduction of H01 gas was continued until there was no appreciable increase in weight of the reaction flask. Two phases separated, the heavier, a dark red oil which was the crude product X, and the lighter, a transparent aqueous layer. The dark oil was separated and stored in a well-stoppered bottle over an- hydrous calcium chloride for at least 24 hours in a re- frigerator. At -10°C. this crude product was a solid. The crude d, -chloro-o(-(p-br0mo,>henyl) -ethane X was distilled in vacuo through an unpacked column of 31 cm. height. No forerun was collected. The fraction boiling 82 - 105°at 1.5 mm. was collected, n50 1.5687 and repre- sented 1005 of theoretical yield. The following table lists the results of a series of runs. _ ., , r‘. 0 Run Gms. of HCl gas Gms. X Yield,% n50 B.P.,C. No. IX used A 100 14 hr. 87.8 80.4 1.5690 107-112/2mm. B 100 13% hr. 101.4 33.2 1.5681 91-104/1mm, c 256.3 15% hr. 223.6 80.0 1.5695 91-97/1mm. D 113.3 14 hr. 123.8 100.0 1.5687 92—105/1.5mm. 14. In all cases the distilled a(-chloro-c(-(p- bromophenyl)-ethane was a water white oily liquid, stable at 20°C. when stored well-stOppered in a re- frigerator for 6 months. PCl was not detected when the bottle was unstoppered occasionally during that period of time. Using the procedure of Cheronis and bntrikin (15), the preparation of the S-alkylisothiourea picrate deriva- tive of cK-chloro-0\-(p-bromophenyl)-ethane was attempted. All fractions of recrystallized material obtained were found to melt above 300°C. No further characterization was attempted. 15. 1,3'DI-(p-BROnDPfiQNYL)-1-BUTENE In the preparation of 1,3-di-(p-bromophenyl)-l- butane, the procedure of Goerner and Hines (1) was followed closely. The course of the reaction may be re- presented by the following equation: 2Br<:j:>0H-CH3 + CuCN H- =CH-CH-CH3 ————> ('31 E: + 2HCN + 20u01(?) Br r X XI In a typical preparation using no solvent, 58 gms. of CuCN was placed in a 500 m1., three-necked flask equipped with a reflux condenser, Hershberg stirrer and thermometer. A glass tube connected the t0p of the con- denser to a coil and receiver packed in ice. A dropping funnel was fitted in the top of the reflux condenser and 142.5 gms. (0.65 moles) of oechloro-OK—(p-bromophenyl)- ethane if placed in the dropping funnel was allowed to drop slowly into the flask. An oil bath surrounding the flask was heated slowly and at 126° mild evolution of hCN began to take place. Approximately 10 minutes after completion of the addition of ct-chloro-OQ-(p-bromophenyl)- etlane, hCN evolution ceased and the reaction subsided. The oil bath was removed and stirring continued for one hour. The cooled reaction mass was extracted with three 16. 150 ml. portions of benzene. The benzene was removed under reduced pressure (as;irator) and the residue was distilled in vacuo through an unpacked column of 31 cm. The following fractions were collected. Fraction Boiling flange at. Gms. Pressure mm. .5 0 A (I - 143 21.5 3 TN /~o r\ c 216 - 225‘ 5.5 3 Fraction B solidified on standing overnight at room temperature. The oily LOlidS were recrystallized from methyl alcohol and 24.1 gms. of a white crystalline solid melting at 67.5 - 68.0'and 38 gms. of a material of lower melting point were isolated. Yield, calculated as XI, wa 47.3% based on recrystallized solids. Analysis of a sample of the solid melting at 67.5- 68° for bromine by the Parr bomb method (16) gave 43.3%, 43.63 and 43.85. Calculated for Clgfilqbrg: Br, 43.6. Stobbe and Posnjak (3) found that halogenation in 032 as a solvent almost completely eliminated side chain substitution reactions for compounds of type XI; for that reason their procedure was used in the addition of bromine to XI. Two and five tenths gms. (.0068 moles) of 1,3-di- (p-brom03henyl)-l-butene was dissolved in C82 in a test tube which was excluded from light by wrapping with a 010th and a 5% solution of BP2 in C82 was added drop- wise to the solution. Bromine discoloration was slow and only enough bromine was added to maintain a brown color in the reaction mixture. One half hour was neces- sary to reach the point of the persistent bromine color. The C82 mixture was poured on a watch glass and allowed to evaporate. The solid residue was taken up in C.P. benzene and the benzene volume reduced to approximate saturation. Petroleum ether was added and a solid separated. Recrystallization from benzene-petroleum ether yielded 3.1 gms. of a white solid melting point 154 - 160°. This correSponded to a yield of 83%. A bromine analysis by the Parr bomb method (16) gave 61.0% and 61.1% compared with a theoretical value of 60.8% for XII. It thus appears that the course of the bromination reaction may be represented by the following equation: hC = CH -CHCH3 Br Br Q + 8P2 HC - CH - ©4343 7" Q 002 Br r B r XI XII Molecular weight determinations on 1,3-di(p—bromo— phenyl)—l-butene were attempted using the procedure of 18. Shriner and Fuson (17) base” on the melting point depression of camphene and pinene hydrochloride. However, results were not reproducible for the depressed melting point of the mixture and so the procedure was abandoned. Freezing point depressions further in benzene as a means of determining molecular weight were attempted, using the procedure of Daniels, Mathews and Williams (18). It was found, however, that at the freezing point of benzene 1,3 di-(p-bromophenyl)-l-butene b gen to display an insolubility which would not permit its use for a re- liable determination. ho other solvents were tried. 1,3-Di-(p-bromophenyl)-l-butene when stored in a stoppered glass vial for a period.of 6 months in a place where incidental sunlight struck it showed a continuous loss in melting point value. It finally became an oily semi-solid material. No investigation was made of this because of the lack of time. Two possible explanations may be offered, however. 1. Possible cis-trans isomerism is potentially existent. One of the isomers may be represented by this preparation and light may supply sufficient energy to convert it to the other. The oily mixture may represent some depressed melting range of mixtures of the two iso- wars. 2. Marion (5) and Stoermer and Kootz (4) 19. observed isomerization of 1,3-diphenyl-1-butene to 1,3-diphenyl-2-butene upon standing with H2804. Light may be sufficient to cause a slow isomerization of the double bond position of 1,3~di-(p-br0mophenyl)-l-butene. A dimerization of o(-chloro—cx-(p—bromophenyl)- ethane was run using redistilled tertiary amylbenzene as a solvent. Using the same conditions described in a typical preparation, 50 gms. of CuCN, 120 gms.(0.548 moles) of .K-chloro-IK-(g-bromophenyl)-ethane and 150 m1. of tertiary amybenzene were used. The reaction took place at O 0 "r'i‘l’ . ,. .- '- “- i . r‘ ° 125 C., but practically no hon was obtained. The reaction . . - O a was malntalned at 138 for two and one hall hours. It was cooled and filtered. The filtrate was stripped of solvent using the reduced pressure of an aspirator. Distillation in vacuo gave the following fractions: Fraction Boiling Range fit. ums. Pressure, mm. . O a 64 - 100 28.6 6 0* ’_ b 100 - 50 10.4 4 c 150 — 205° 32.3 2 *Major portion received at 128°/ 4mm. Fraction 0 represented the crude yield of dimer X . Yield 29.6%. From this data the use of a solvent decreases the yield of dimer. hack of time prevented a further investigation of this product. 20. DEGHALATION STUDIES ON l,3-DI(p-ERO¢DPHEHYL)-l-BUTENE A. Hydroxylation. The procedure of Miles and Sussman (19) was used in the hydroxylation of the double bond of 1,3-di-(p- bromophenyl)-l-butene X . The course of the reaction may be represented by the following equation. HOH 0% H 3 C- C- CH3 - C - CH3 ‘ + H202+~ 0304—; PEO: ~ t-BUOH Br XI XII Preparation of the hydrogen peroxide reagent was accomplished in the following manner (l9). Four hundred milliliters of redistilled tertiary butyl alcohol was added to 100 ml. of 30% hydrogen peroxide and to the so- lution was added small portions of anhydrous sodium sul- phate until two layers appeared. The tertiary butyl alcohol containing most of the hydrogen peroxide was separated and dried over anhydrous sodium sulphate and finally stored over anhydrous calcium sulphate (Drierite) in the cold at approximately 10°C. A solution of approx- imately 6-7% H302 was obtained. If all of the water were not separated properly, the H202 content decreased slowly on storage. If, hOWever, all water was removed the solution of R202 in tertiary butyl alcohol was stable for periods up to six months. Osmium tetroxide dissolves readily in tertiary butyl alcohol and is perfectly stable provided no iso- butylene is present. Otherwise, it is reduced to an insoluble black oxide which can be oxidized by H202. When a 0.5% solution of osmium tetroxide in t-butyl alcohol was prepared and stored in the light, the characteristic black precipitate formed. It was found that the addition of a small amount of H202 in tertiary butyl alcohol caused the black precipitate to dissolve and restored the catalyst to an active condition. In the actual hydroxylation of 1,3-di-(p-bromo- phenyl)-l-butene, an excess of 0504 was used in contrast to 1 ml. of catalyst solution recommended for each 0.1 mole of double bond. For a typical reaction, 2 ml. of 0.5%0s04 solution in tertiary butyl alcohol was used for 0.0274 moles of double bond. Further, nine-tenths of the theoretical amount of H202 was found to permit fragmentation of the glycol XII possibly by avoiding an excess of reagent which might cause extensive oxida- tion. Ten grams (0.0274 moles) of l,3-di-(p-br0moghenyl)- [‘0 [\3 . l-butene was dissolved in 175 ml. of tertiary butyl alcohol. Two m1. of 0.5% 0s04 solution was added at room temperature. ho temperature change was noted. Ten and eleven hundredths gms. (14.5 ml.), (0.0246 moles) of a 7.36% h202 solution was added in increments, first 5 m1., 15 minutes later, 5 m1., and ten minutes after the second addition, 4.5 ml. The reaction was very mildly exother- mic with no more than a three degree temperature rise of the mixture noted in any run. The reaction was then all- owed to stand until the next day with partial exclusion of light. B. Distillation Previous to diStillation the hydroxyla- tion reaction mixture was treated with 2 gms. of zinc dust and 3 ml. of water to destroy any h202 which might still have been present. Distillation was carried out in vacuo in a 50 ml. 24/40 standard taper flask equipped with a finger condenser type distilling head. Solvents were first removed at room temperature with a water aspirator. 23. The following fractions were collected: Physical Appear- Fraction Boiling Range Pressure nEO ance at Room L Temperature 1 76 - 260 2 mm. - solid 2 89 - 90° 2 mm. - H 3 97 -107” 1% mm. 1.5760 green liquid 4 115° 2 mm. 1.5760 " " 5 125 -1470 1% mm. - " " 6 160 -1920 2 mm. - liquid slowly solidfying Fractions 1, 2, 3, 4 and 5 gave an orange precipi- tate with a 5% solution of 2,4-dinitrophenylhydrazine in 20% perchloric acid. Previous runs which had been treated with ceric per- chlorate to cleave the glycol gave results similar to those recorded above. Thus, because the heat of distilla- tion appeared to be sufficient to cleave the glycol XII. it was felt unnecessary to continue the ceric perchlor- ate procedure. C. Characterization of Fractions. 2,4-Dinitrophenylhydrazones were prepared for fractions 1 and 3. The procedure of Cheronis and Entrikin (20) was used. 24. a small quantity of fraction 1 was dissolved in 10 ml. of methyl alcohol and added to 2L0 ml. of methyl alcohol saturated with 2,4-dinitr0phenylhydra- zine. mild reflux was allowed for about 8 minutes and several drops of concentrated ECl were added to the point of incipient precipitation. Refluxing was continued five minutes and 2 ml. of water was ad ed. The insoluble hydra- zone precipitated. After the solution stood overnight at room temperature, the derivative was filtered, washed with water and some methyl alcohol and dried, m.p. 253-254 . Recrystallization from methyl alcohol raised the melting point to 257-25 0. A mixed melting point with an authentic sample of the 2,4-dinitrophenylhydrazone of p-bromobenzalde- hyde, reported to melt at 260-26lo(21), (see section 0n heference Aldehydes), melted at 2570. Fraction 5 was treated by the same procedure as described for fraction 1. After the first crop of crystals, A was filtered, water was added dropwise and a second Similar treatment gave C. The addi- The crop B was obtained. tion of excess water caused crop D to precipitate. melting points of the fractions were: 0 £53.. 235 _ 240 C. 19000. '13 '._n 0') 0'1 I c. 118 - 121°C. n. 109 - 195°C. 5‘) Ch 0 PHSIARATION'CF asrsasn"m Antnnruss In the early phases of this investigation, the proof of structure of l,3-di(p-bromophenyl)-l-butene XI was contemplated by the characterization of 2,4-dinitro- phenylhydrazone derivatives of carbonyl components ob- tained by fragmentation of the hydroxylated butene XII. p-Bromobenzaldehyde had been previously prepared (13) and heilbron (19) reported the melting point of its 2,4-dinitrOphenylhydrazone as 128°. This did not compare with the derivative isolated in our reaction and was later found to be in error (14). Further, an error of 10 degrees in the melting point of p-bromobenzaldehyde was found in Coleman and honeywell's procedure (13) for its preparation. p-Bromohydratropaldehyde had not been reported or characterized in the literature previously. For these reasons, it was decided to prepare these aldehydes and characterize them in order that they might be used for comparison purposes with those isolated in the degradation studies of l,3-di(p-bromophenyl)-l- butene. , PARA BRQH b HZALbLPYDE p-Bromobenzaldehyde was prepared according to the procedure of Coleman and Eoneywell (22). The course of the reaction may be represented by the following equations: BrQCI-13+ 2Br2 -———) BrOChBr2+ 21Br x111 XIV BI‘QCHBI‘B + 1120 -———-.s BrOCIZO + 2am XIV XV In a 1-1. three-necked flask equipped with mechani- cal stirrer, reflux condenser, thermometer and dropping funnel was placed 200 gms. (1.16 moles) p-bromotoluene which was commercially available. Flask and contents were illuminated with an unfrosted 150 watt lamp and heating of the contents was commenced. Addition of 394 gms. (2.46 moles) of bromine was started at such a ate that there was not a large excess present at any time. The bromine was introduced under the surface of the reaction mixture. About one-half of the bromine was added in the first hour at 105-110°C. The rest was added over about 3 hours at a temperature of 135°C. The temperature was then raised slowly to 150°C. The chilled crude p-bromobenzal bromide was trans- ferred to a 5 1. round bottom flask and was mixed with 400 gms. of CaCOg. Six.hundred milliters of water was ad‘ed and the mixture was heated under reflux for twenty- six hours to effect hydrolysis. The hydrolyzed mixture was then steam distilled, insuring complete agitation of the heavy solids, through a Hopkins still head to minimize bumping over into the condenser. The first 3 1. contained most of the product but an additional 3 l. of distillate was collected. Yield: 148.5 gms. h.P. 67-68° 24.6 gms. 63-64° total 173.1 gms. 78% yield The 2,4-di-nitrOphenylhydrazone (XVa) of p-bromo- benzaldehyde XV was prepared by the method of Cheronis and Entrikin (20) previously described. The melting point was 257-258° after one recrystallization from methyl alcohol. Snyder and Kendrick (21) report the melting point of the 2,4-dinitr0phenylhydrazone of p-bromobenz- aldehyde as 260-1 . R" Q); 0 PARA BROEJOHILRATHC/r ‘ 1 I .41: YES The Darzens glycidic ester condensation was used for the synthesis of p-bromohydratropaldehyde. p-sromo- acetophenone XVI and ethyl chloroacetate XVII are com- mercially available so this Qp ea red to be the shortest route. Allen and Van Allen (23) had previously converted ethyl- fl -methyl- (3 -phenyl-glycidate to hydratrOpaldehyde in 50% yield and Newman and Closson (24) had carried out the same reaction with minor modifications of conditions. The course of the reaction may be represented by the following equations: 0 BrQE-CH3+ ClCHZCOOEt ———-> Br é-‘ca—coont Nalmg CH3 XVI XVII XVIII Br c’O‘CH 000% * H Br© ’O‘CF CO0Na + tOH - — In - 1- 11: XVIII “ XIX ’0‘ Br c-CH-COONa —-—3—> Br©¢-cao XIX XX Sodamide was prepared by the procedure of Hancock and 30. Cope (25). The preparation was carried out in a well ventilated hood. A 1-1. three-necked flask equipped with a reflux condenser, which was stoppered with a soda lime drying tube and a glass inlet tube’for introducing ammonia was chilled in an acetone-carbon dioxide bath. Ammonia from a cylinder was introduced slowly to allow liquifac- tion of the gas. When approximately 500 m1. of liquid Nhghad been collected the flask was removed, a crystal of Fe(N03)3 added, followed by 92 gms. (4 moles) of sodium metal cut in small pieces. Stirring was continued through- out the addition. When the liquid NH3 had evaporated the solid sodamide was removed and stored under dry nitrogen in wax sealed rubber stoppered bottles. Yield 138. gms., or 87.1%. Ethylfl -(p-bromophenyl)-/3 -methy1glycidate XVIII was next prepared using sodamide as the condensing agent. One hundred grams (0.5 moles ) of p-bromoacetophenone XVI, 62 gms.(0.5 moles) of ethyl chloroacetate XVII and 150 ml. of anhydrous C.p. benzene were mixed in a 2 1. three- necked flask equipped with thermometer, mechanical stirrer and a reflux condenser carrying a soda lime drying tube. Twenty-three and six-tenths grams (0.6 moles) of soaa- mide, which was well powdered, was added in small incre- ments over a two hour period maintaining a reaction 31. temperature of 15-2000. by means of a crushed ice bath. Stirring was continued two hours after completion of soaamide addition. The contents of the flask were poured into 350 gms. of crushed ice and the benzene layer separated. The aqueous layer was extracted twice with 200 ml. of benzene. The combined extractions were washed three times with 100 ml. portions of water, the last portion containing 10 ml. of glacial acetic acid. The washed extract was dried over anhydrous sodium sulfate overnight. The benzene solution was transferred to a 150 ml. 24/40 standard tapered, round bottom flask and the ben- zene evaporated at room temperature under the reduced pressure of an aspirator using a Claisen head. For the distillation of the glycidic ester XVIII a distilling head of the finger condenser type was used. Distilla- tion was carried out in vacuo using a Cenco LiVac pump. The following fractions were obtained: LISTILLATE Fraction Wt. gms. Boiling Range Pressure age A 21.73 38 -113'0. 3 mm. solid B - 113” 25 mm. H C - 113-120" 2; mm. . n D 6.8 125-127" 2; mn. 1.5530 10.7 130-136" 2 mm. 1.5400 E F 18.0 137’ 13 n. 1.5361 G 17.1 138-140° lg-mm. 1.5360 H 7.6 l40-140.5° 1% mm. 1.5372 I 6.4 141-146” 2 mm. 1.5380 J 0.7 to 150° 2 mm. 1.5414 a)-comb1ned weight of fractions A,B and C. Analysis of fraction G using the Parr bomb method (16) gave 28.0% bromine compared with a theoretical value of 28.0% for ethyl p -(p-bromophenyl)-fl -methyl— glycidate XVIII. This glycidic ester XVIII appeared to be quite stable when stored in a well stoppered vial for periods of one to two months. It is a straw-colored, viscous oil, b.p. 139°C. at 1.5 mm nfiO 1.5360. The yield was 30.6% based on fractions F through I. Fractions A, B and C consist mostly of unreacted p-bromoacetophenone and ethyl chloroacetate. This is characteristic of previous reports of the condensation reaction for preparation of other glycidic esters. Hydrolysis and decarboxylation of the glycidic ester XVIII essentially followed the procedure of Allen and Van Allen (23). The p-bromohydratrOpalde- hyde XX was distilled in vacuo, however, because of its 33. probable liquid character. To 25.6 gms. (.0885 moles) of ethyl.fi3-(p-bromo- Phenyl)-(gamethylglycidate (from vials F and H), con- tained in 1-1. three-necked flask equipped with mechani- cal stirrer, reflux condenser and thermometer, was added a solution of 8.4 gms. (.210 moles) of sodium hydroxide in 24 ml. of water. Almost immediately an exothermic reaction set in and the temperature of the reaction mixture rose to 65°C. The flask was chilled and the temperature maintained at 45-5000. by means of a water- ice bath. An insoluble white solid precipitated. After stirring for five hours, the reaction mixture was acidified with 6 N. HCl until blue to Congo paper. External cooling of the flask was necesSary. The acidi- fied mixture was refluxed for two and one half hours and upon cooling, an oily layer separated. The water layer was extracted with two 30 ml. benzene portions and the combined oily layer and extracts placed under the re- duced pressure of an aspirator to remove the benzene. The residue was distilled in vacuo using a finger condenser type distilling head. The following fractions were collected: 34. . n . a. . . . £0 Eractlon at. wms. BOlllng Range Pressure n2 1 2.2 90- 96°C. 2 mm. 1.5604 2 6.2 98-100°C. 2 mm. 1.5602 3 2.3 loe-llo°c. 2 mm. 1.5602 4 1.3 117-127°C. 2 mm. 1.5605 5 1.2 lb‘O-l46°C.a 2 mm. 1.5623 6 1.8 147—165°C.a 2 mm. 1.5701 a) Decomposition noted during distillation Fractions 1-4 represent a 65.7% yield of aldehyde XX from the glycidic ester XVIII. The distillate was a green-white liquid in the first four fractions. A drop of fractions 1 and 3 gave precipitates when mixed with 5 ml. of a 5% phenylhydrazine solution in 20% perchloric acid. Bromine analysis of fractions 1 and 3 by the Parr bomb method (16) gave respectively, 39.6% and 39. % as compared with a theoretical value of 43. % bromine for p-bromohydratropaldehyde XX. If XX may be compared with hydratropaldehyde, its stability would be expected to be very poor because of susceptibility to oxidative influences. This would explain the poor comparison of halogen analysis with the theoretical. A 2,4-dinitrophenylhydrazone XXa of p-bromohydra- tropaldehyde was prepared by the method of Cheronis 0) CT! 0 and Entrikin (20) and recrystallized from glacial acetic acid to remove excess 2,4-dinitrophenyl- hydrazine. A bromine analysis of this phenylhydra- zone (XXa) (16) indicated 20.1 and 20.5% bromine compared with 20.3% theoretical bromine. p-Bromohydratropaldehyde-2,4-dinitrophenyl- hydrazone (XXa) H N0 :Br<::>-C-CH=N-h-i::>hol e h Ha v" ~ r an is a yellow crystalline solid melting 123-124°C. It crystallizes as needles from methyl alcohol and prisms from glacial acetic acid. Its comparative solubility in methyl alcohol is greater than p-bromo- benzaldehyde-2,é-dinitrOphenylhydrazone (Xva). XXa was mixed with the 2,4-dinitropheny1hydra- zone of crop (C) (see page 26) and a melting point of 118-119° was observed. A bit of crop (C) was seeded in a saturated solution of XXa in glacial acetic acid and crystallization took place. The dried crystals melted at 123-124°. Thus crop (C) and XXa must be identical and the second component of the hydroxyla- tion degradation is p-bromohydratropalcehyde. 36. DISCUSSION The work of Goerner and Eines (1) has been extended 'to 0(-chloro-a(-(p-bromophenyl)-ethane X which was pre- pared from p-bromophenylmethylcarbinol IX (see Experi- mental for preparation). It is interesting to note, that o(-chloro-fl%(p-bromophenyl)-ethane, prepared in the manner described (page 14) gave average yields of 10% higher than that reported by Barber et al. (11) who used thionyl chloride instead of HCl for convertlng the alcohol to the chloride. Further, Barber reported o(-chloro-°F(p- bromophenyl)-ethane in what would seem to be its more crude state. we have found it to be a pinkish pungent oil in the crude state, whereas upon distillation, this crude product gave a water white 011 of physical proper- ties comparable to those described by Barber (11). It seems highly probable that analysis of the aliphatic chlorine in 0(-chloro-°<-(p-bromophenyl)- ethane is feasible by a sodium hydroxide reflux technique. he were able to titrate an aliquot which had been re- fluxed for twenty-six hours in 6 molar excess of 6N. alcoholic aqueous NaOH and obtained results only within 1% of the calculated values. It was felt that question- able purity of the original sample might add to the 57. explanation of low results. The reaCLion of DK-chloro-‘Kip-bromophenyl)- ethane with cuprous cyanide can be compared favorably with the PBECUIUD ofC(-chloroethylbenzene. Goerner and Hines reported yields of 40% for their dimer. In this investigation we have obtained yields of dimer of the same order of magnitude. It would appear that this dimerization reaction is characteristic of C(-halo- ethylbenzenes. The reaction does not, however, seem to be as violent in this case as that observed by Goerner and Hines (1). It starts at approximately the same temperature but there is not the violence of filling the condenser with ECN or its rapid generation noted in the previous instance. This may be partially explained through the fact that molecular ratios of CuCN and.o(- chloro-O(-(p-bromopheny1)-ethane do not require as much CuCN as was needed in the case of cK-chloro-‘K—phenyl ethane. Thus the actual generation of hCN on a weight basis would be much less in the preparation of 1,3-di(p- bromophenyl)-1-butene. The structure of l,3-di(p-bromopheny1)-l-butene seems to be reasonably well established. when sharply melting crystals were hydroxylated and cleaved either by heat or ceric perchlorate and heat, the carbonyl components were isolated and characterized as their 2,4-dinitr0phenylhydrazones. p-Bromobenzaldehyde had previously been prepared and characterized as the 2,4-dinitrophenylhydrazone. It was easily isolated in the distillate of a hydroxy- lation degradation and precipitated as the 2,4-dinitro- phenylhydrazone. 0n comparison with an authentic sample of p-bromobenzaldehyde-2,4-dinitrophenylhydrazone there was no melting point depression. Isolation of p-bromohydratropaldehyde as a component of the hydroxylation fragmentation was more difficult. It had not been previously reported in the literature. gre- paration through the Larzens glycidic ester condensation was accomplished during this investigation. It was characterized as the 2,4—dinitrophenylhydrazone. Once p-bromohydratropaldehyde had been prepared and characterized, sufficient information was available to choose distillate fractions of the hydroxylation frag- mentation which were richer in this component. By pre- ciaitation as the 2,4-dinitr03henylhydrazone and frac- tional crystallization from methyl alcohol p-bromo- benzaldehyde-2,4—dinitrophenylhydrazone was isolated from the hydroxylation fragmentation mixture and com- pared by mixed melting point with the prepared sample. ho depression was noted. having established the structure of 1,3-di(p- bromophenyl)-l-butene, bromination in 082, which is an accepted method of adding to the double bond, undoubted- ly yields l,2-dibromo-l,5-di(p-bromophenyl)-butane. halogen analysis compared with calculated values for the brominated butane derivative seems to bear this out. Little accurate information is actually known about the mechanism of this dimerization reaction. Goerner (2) has observed that the use of styrene as a solvent in the dimerization of cxechloro-OK-phenylethane greatly increases the yield of dimer. If sufficient styrene were present,yields greater than 100% of theoretical based on dimerization (eq. see p. 16) were obtained. Further, p-bromostyrene must be present in the dimerization re- action of ox-chloro-ok-(p-bromophenyl) ethane. when a forerun of the distillation was refluxed with a saturated solution of maleic anhydride in benzene and a trace of benzoylperoxide, in about ten minutes a polymeric preci- pitate formed. It is a well known fact that OK—chloroethylbenzene decomposes to form styrene. By analogy we could expect the following reaction: ’“1155 -- , h H ’C1 HCl + 8H2 —> a.) Br Br- 40. p-Bromostyrene may exist in at least the indicated possible resonance forms. (SIP-'2 (4(ng CH 5 CH b-) 6 (+8) Br If reaction (a.) takes place a possible intermediate step in the formation of p-bromostyrene might be the for- mation of a carbonium ion. " r ‘l pus Chg H—C-Cl H- (+) g..— 0.) Cl‘ Br Br Reaction of the carbonium ion in (c.) and the resonance structure given in (b.) would give a dimer 1" 9H3 fl <->§H2 _ me (+) C'H CHj-CH-Ch’g- HT .. ( C1 +( ———> 0 (:1- d.) L_Br _J Br [_ Br 5r In the presence of a molecule of CuCN a molecule of hCl would tend to split out of the ionic form of the dimer accompanied by rearrangement. e.) CH3- -CH2- H + HCl Br In an attempt to convert the dimer from the gig to the trans forms or the reverse, 1 gram of l,3di(p-bromo- phenyl)-l-butene was dissolved in 25 ml. of nitrobenzene and refluxed for twenty-six hours with a trace of iodine (11). A polymeric material was obtained. This would further help to exglain the tendency of the double bond in the dimer to undergo addition type of polymerization. 42. SURMAHY l. o(- Chloro-¢(-(p—bromophenyl)-ethane has been prepared and compares favorably with the preparation reported by Barber et al.(ll). 2. o(-Chloro-0<-(P-bromophenyl)-ethane has been dimerized to l,3-di(p-brom0phenyl)-l-butene by the treat- ment with CuCN in yields of the order of 47%. 3. The structure of l,3-di(p-bromophenyl)-l-butene has been established by hydroxylation and fragmentation of the aliphatic double bond to carbonyl components. 4. The carbonyl components, p-bromobenzaldehyde and p-bromohydratropaldehyde, were characterized as their 2,4-dinitrophenylhydrazones and compared with prepared samples of each. 5. p-Bromobenzaldehyde was prepared and characteri- zed as previously reported in the literature (20) (22). 6. p-BromohydratrOpaldehyde had not been previously reported in the literature. It was prepared through a Darzens glycidic ester condensation and characterized as the 2,4-dinitrophenylhydrazone. 7. l,3-Di(p-brom0phenyl)-l-butene was brominated to 1,2 dibromo-l,3-di(p-bromophenyl)-butane. l. 2. 3. 5. 6. 7. 8. 9. 10. ll. 12. 18. 14. 15. l6. 17. REFERENCES Goerner and Hines, J. Am. Chem. 80c., 70, 3511 (1948). Goerner, Private observations. Egloff, "Physical Constants of Hydrocarbons” Vol. III Reinhold Publishing Corporation, hew'York, N.Y., 1946, p. 398. Stoermer and Kootz, Ber., 61, 2330 (1928). ( Marion, Can. J. Research, 16 B, 213 1938). Fittig and Erdman, Ann., 216,187 (1883). Erlenmeyer, hnn., 135, 132 (1865). Swartz, Ann., 132, 231 (1866). Stobbe and Posnjak, Ann., 371, 284 (1909). Kbnigs and Mai, Ber., 35,2658 (1892). Barber, black and Woolman, J. Chem. 80c., 99 (1943). Ziegler and Tiemann, Ber., 65, 3414 (1922). Quelet, Bull. Soc. Chem. (4), 45, 86 (1926). Overberger, Saunders, Allen and Gander in bnyder,"0rgan— ic Syntheses," Vol. 28, John Wiley and Sons, Inc., new York, N.Y., 1948, p.30. Cheronis and Entrikin, "Semimicro Qualitative Organic Analysis” Thomas Y. Crowell Co., New York, N.Y., 194/, p. 282. Fisher, "Laboratory Manual of Organic Chemistry", r “1 John Wiley and bons, Inc., New‘lork, N.Y., 1931 ad., p. 344. Shriner and Fuson, "Identification of Organic Compounds” 1940, p. 122. 44. 18. 19. 20. 21. 22. 23. 24. 25. Daniels, mathews and hilliams, "i‘perimental Physical Chemistry", 3rd. md., McCraw-hill Book 00., New'York, N.Y., 1941, p.76. hilas and Sussman, J. Am. Chem. Soc., 68, 1302 (1936). See Ref. 15, p. 247. bnyder and Handrick, J. Am. Chem. Soc., 66, 1860 (1944). Coleman and Loneywell in Blatt, "Organic Syntheses Collective Volume 2”, John Wiley and Sons, Inc., New York, h.Y., 1948, p.89. Allen and Van Allen in brake, "Organic Synthesesfl, Vol. 24, John Wiley and Sons, Inc., new iork, N.I., 1944, p.82. Newman and Closson, J. Am. Chem. 800., 66, 1553 (1944). hancock and Cope in Eachmann, "Organic Synthesest, Vol. 25, John Riley and Sons, Inc., new Yorx, n.1., 1945, p. 25. 45-