1‘ | l‘ H 1 W I WW I l I I , +| THS' A STUDY C??? THE REACHGK OF MEI-Wt. ETHYL ésopmpvi {ZARBEHGL WITH ‘5EMZENE ABE ALUMINUM CHLCDRéD-E ‘Fhasis *Ear fié'ia- Daeg'me M M. 3: MECMGAM STATE {EQLLEQE ‘Wil‘éiam 5. Thurber 194.3 THEE-11;” ~ ‘ is. fl-F— [' I, ‘ ‘\ . I 3 I '\ .I 1 » 3‘4 n43.53'n 5 {Ci '1 a“; A STUDY’OP THE REACTION OF NETHYL ETHIL ISOPROPYL CARBINOL WITH BENZENE AND ALUHINUM CHLORIDE by WILLIAM S. THURBER A TEESIS Submitted to the School of Graduate Studies of lichlgan Stnta College of Agriculture and Applied Science in partial fulfilmlmut or the requirement- for the degree of MASTER OF SCIENCE Department of Chemistry 19h8 ACKNOWLEDGMENT The writer wishes to express his sincere appre-v ciation to Dr. Ralph 0.. Huston for his aid and guidanoe through the course of this investigation. 331-671 CONTEHTS Page Introduction........................................ 1 Hiatorical.......................................... 3 Theoretical......................................... 6 Experimental........................................ 10 1. Preparation or’nothyl Ethyl Ilopropyl Carbinol............................. 10 II. The ommatim Appamtnl....nuuuun 11 III. condensation, Separation and Fraction- ‘tionoocooooooocoocootocococoa-cocoon 13 1?; Identification 0! Fraction................ 1? D1!¢U3‘1°floso.coo-o.coo-cocooconccocooccooococoon... 23 SWoocootoocoooooccoooococoooocoooccoococoon-coo 28 Bibliographyooooaocean-ooococoocooococco0.000.000... 29 INTRQDUCTION The condensation of aliphatic tertiary alcohols with benzene or phenol in the presence of anhydrous elm- inun.chloride has been studied extensively by Huston and co-workerc in this laboratory. In the early work, beginning with that of Bus- ton and Priedenann (1) in 1916, the studies were primarily concerned with the formation or the direct condenration product. However, all workers have reported fragmentation to some degree. This fragmentation being greeter’with highly branched alcohols, with a subsequent lowering of yield: in the expected tertiary alkylbenzenee. Recent studies (2) (3) (h) (5) have been con- cerned with the products of fragmentation and rearrange- ment when reactive conditions and molar quantities of the reactants were varied and when branching on the carbon atom adjacent to the hydroxy carbon was increased. Huston and Binder (6) have condensed the heptyl alcohol: with benzene and obtained good yields of the corresponding hydrocarbon from the straight chain carbin— ole. with the branched chain alcohols, the hydrocarbon was obtained in small yield with considerable anounta of the halogen derivative of the carbinol being formed. lo other products were definitely isolated. It is the purpose of this investigation to con- tinue the study of the reaction of methyl ethyl isopropyl carbinol with‘henzene in the presence of anhydrous aluminum chloride in an effort to determine what products of frag- mentation and rearrangement are produced. HISTORICAL The studies of the condensetions of alcohols with benzene in.ths presence of anhydrous alundnnn.chlcr~ ide were begun in this laboratory in 1916, when Huston end Priedemenn, repeating the work of let (7), condensed bensyl alcohol with benscne in the presence of aluminum chloride. In addition to diphenylnethsne obtained by Her, these work- ers reported the formation of die and tri-substituted benzencs. Several primary alcohols, methyl, ethyl, propyl, isopropyl, hntyl, isobutyl and the isoanyl alcohols, were condensed by Huston and Sager (8), using one-halt’nols of aluminum chloride to one mole of alcohol. The expected elkylbenzene use not obtained. In 1933 Huston and Davis (9) condensed triphenyl carbinol with benzene and obtained triphenylnethane rsther than the expected tetraphenylmethane. Huston and Goods— moot (10) have condensed three cycloalkyl carbinols to chew that the east of condensation was related to the strain in the ring, a greater strain being more conducive to condensation. 'Tzukervsnik (ll) round that saturated primary aliphatic alcohols do not condense, simple sec- ondary alcohols give a 25-30 per cent yield of the ssca ondary alkylbensene and tertiary butyl alcohol gives a 60~70 oer cent yield of trinethylphonylmethene. Huston and others (12) have condensed tertiary butyl, tertiary anyl, the three tertiary hexyl, seven tertiary'heptyl and all of the tertiary octyl carbinola with benzene. From.theee condensationa it has been found that the mole ratio of alcohol to aluminum chloride is one to one-half respectively for the optimum production of the corresponding none alkyl benzene. An increase in branching on the alpha carbon decreases the yields of the expected alkyl benzene and increases the formation of unsaturated compounds, halides and lower alkyl benzenes. Condensations at lower temperatures decreases the amounts or these side products. Huston and hughes (l3) condensing sense dialkyl- aryl carbinols with benzene have reported formation of diners or the corresponding unsaturated products of the alcohols, as well as the expected diphenyl alkanss. Kaye and others (1%) have condensed a number or secondary alcohols with benzene and round that in gear oral, more than .3 moles or aluminum chloride per one- half mole of alcohol decreases the yields of the mono- alkylbenzenec. The addition of dry’HCl throughout the addition or alcohol increases the yields. welsh and Drake (15) have condensed several tertiary aryl carbinols with benzene and have reported the formation or the correspond» ing alkene of the carbinol, as well as their saturated dimers. Norris and co—workers (16) have reported the formation of symmetrical tri—alkylbenzenes from the con» deneation or primary aldehols. Huston and Barrett (2) have studied the cone densation or the dehydration product of tertiary butyl carbinol and have shown that the reaction is greatly aided by the addition of H61. The condensation of the alkene however did not give the chloride or the alcohol as does the condensation of the alcohol itself. By mixing tertiary heptylbcnzene with aluminum chloride they obtained split products indicating the reversibility of the skylation reaction. Huston and van Dyke (l?) have found a pronounced denethylation in the condensation of several more highly branched emu... In recent studies by Huston and Smith (3) it was found that by increasing the proportion of anhydrous aluminum chloride from one~third to one-half mole per’nole or alcohol in the condensation of t-methyi-h-heptanol the skylbensene fraction was increased approximately fifty per cent. They have also shown that reduction may take place during the condensation reaction and cite as evi- dence the isolation of the paraffin hydrocarbon fraction frcn.the condensation of non-branched alcohols and iso- butane from 2,t,t trimethyl-zupcntanol. THEORETICAL The isolation of elkenec and alkyl halides from the condensation of aliphatic alcohols with benzene in the presence of anhydrous aluminum chloride show that akylntion of benzene nith alcohol must be closely related to the alkylaticn of benzene with alkenec and alkyl hal- idee. Good yields of the elkylbenzenc in both caeee is additional evidence of this relationship. Txukervanik (11) has proposed a mechanism for the akylation of benzene with aliphatic tertiary alcohol- uning an a basis the formation of the alkyl halide an an intermediate. t~C53110H + A1013 ———————9 t¢053110A1C12 + ROI tv-C53110111012 ————9 05310 + A1012 OH 05310 + H01 —-—O t-‘Csflllcl t~05311C1 + 05115 “1°13; “563110635 + an The isolation or elkenec and alkyl halides as bybproducta has been offered as experimental evidence to support such a mechanism. Tankervnnik suggests that the evolution or H01 was due to the formation or the eluminnn.chloride - alcohol complex with n subsequent elimination of the gel. Th1: some mechanism ha- egain been proposed by Tzukervanik to explain the condensation of secondary al- cohols with benzene. However, no chlorides or nlkcnee were 7 found as hwhproducts from these condensations. This type of lechsnisl hac failed to explain the rearranged prod- not. found by Kaye (lit), in condensing some secondary clcoholl with benzene in the preconcc of anhydrous slun- inum chloride. Norris and Sturgis (16) have proposed a similar mechanicn with essentially the same stepc involved. lcxennc and Son (18) have proposed a mechanism whereby a resrrlngensnt of the alkyl groups teln- place on certain primary and secondary alcohols. lhic is based on the reflection.of alksncc during the reaction. Ubing Boron tri-fluoride as a catalyst they have isolated small amounts or the alkenes and their polymers. a-cuz-cuzon BF?! . R-CH:CH2 ——T n-cn-cng + 05115 ———-. a~cn(csn5)~csg By this mechanism the condensation of a primary alcohol leads to a secondary alkylbensene. Price (19) has proposed an ionic mechanism in- volvinge catanoid attack on the ring. Conductance stud- ies by Ucrtyporoch and Pirla (20) have shown the existence or an ionic complex between aluminum chloride and an al- kyl halide. mag: + gist”. = H" + (:Xiézcn‘ C Ulioh and Home (21) in studying the equilibrium for the formation or certain catalyst‘alkyl halide complexes postulated by Uertyporouch found that the rate of alkylsa tion is directly proportional to the concentration or the cstalyst-alkyl halide complex. This same type or complex may be shown using alcohols or other oxygen containing compounds such as ethera or esters. C1 . X + - mg: 9 3:01: 3:8:é :01 = R + (30111013) According to Price. the electron deficient car- boniun ion romed above may replace a hydrogen am in benzcnoid compounds by association with a pair of electrons from a double bond or the aromatic nucleus to complete its octet. mg: + 0 -———1 023§xn ____, 00-]! + 3" This reaction becomes more complex in the more highly branched tertiary alcohols. The isolation of such pro- ducts as tertiarybutyl benzene is an indication that frag- mentation and rearrangement nmst take place as a side reaction. This side reaction may be explained by a shift or an alkyl group to the positive carbon with a shift in the electron arrangement of the molecule so that the car- bon adjacent to the original positive carbon now becomes positive with a resulting scission or the chain. l..:§n¢H;; ———) 63 “a f £3333_—> GIG-g; + + 032.0112 9. vi, Th1. rearrangement and cleavage allows the activated ben- zene nucleus to react to form tertiary butyl benzene in the case selected. 633-5533 + O : 0- 3326113 6113 This explanation 1: in accordance with mum-am (22) view on molecular rearrangement- and u111 easily explain the rearrangement: observed by Kaye (14) in thia labor-au- tor-y. lo EXPERE'ENTAL I. ngaretion 9;; Methyl Ethyl Ieopropxl Cerbinol This alcohol was prepared by use of the Orig- nerd reaction, by the addition of Methyl isopropyl ketone to ethyl magnesium bromide, followed by hydrolysis of the resulting addition compound. The procedure folloled was that used by c. H. Hedrick (23) in this laboratory. The Grignard reagent was prepared by placing 36 g. (1.5 mole.) of dry magnesium turningc end 200 cc. of anhydrous ether in a one-liter, three-necked flask equipped with reflux condenser, dropping funnel and a mercury seal mechanical stirrer. To this mixture was added 153.5 g. (1.5 moles) of dry, redietilled ethyl bromide in ice :1. of Mydroul ether. The mixture was stirred during addition and the ethyl bromide and ether added at a rate Just fast enough to keep the mixture re» {lazing gently. After cdditim was complete, ltirrihs was continued for me hour at roan temperature. At the end or this time the Grignerd reagent was titrated accord- ing to the method developed by Gilman, Wilkinson, Fiechel, and heyere (24). The calculated amount or arignard nu 1.4 mole: or e 9" per cent yield. 0n the basic of this yield, 122 g. (1.}; holes) of redietilled methyl ieopropyl ketone in 200' cc. of anhydrous other was added dropwiu and stirring nan continued for two hours at room temper- ature. me mixture was cooled in an ice bath and then poured into a large beaker of ice. Dilute hydrochloric 11 acid was added to dissolve the magnesium hydroxide. The other layer was separated and the water layer extracted three times with ether. The combined other layers were then washed with a ten per cent solution of sodium car- bonata until neutral to litme and then dried for several hour: over anhydrous aodiun carbonate . 'me ether waa reacted through a packed calm and the remaining alcohol waa diatilled and collected over a two-degree range at reduced preuure. Yield: 51% 3913 3 338-4900 fine 1 138-1uo°c he: 3 1.11292 l‘hia alcohol has been prepared by Whither. and Ever. (25) and by hedriak (23). Physical constanta ob- tained check with thou Hhitnorc and more obtained by preparing the alcohol from the reaction or nethyl ethyl ketonc and icopropyl magnesium bromide. II. The @ndencatign Apparatus The condensation apparatua is essentially the cane aa that developed by Van Dyke (17), variations of which have been and extensively in this laboratory. Figure l ahcwc a diagram of the apparatus and the coun- tial parts are listed below: A. A one liter, three-necked, round bottom flash. 3. Dropping humel with inner acaled tube and D. E. F. O. H. I. J. K. L. I. 12 connections for equalizing the pressure to form an air-tight system. Mechanical stirrer consisting of an electric motor and glycerine sealed glass stirrer. Carbon dioxide generator system consisting of a vacuum bottle for solid carbon dioxide connected to a each bottle consisting of concentrated sul- furic acid and a mercury safety valve. A side arm delivery tube to which is attached a thermometer and reflux condenser. A two-way stopcock used to by«pass the ice-salt cooled trap. Iceoeelt cooled trap to condense the vapors of the more volatile liquids, such as benzene and alcohol. A two-way stopcock used to by-pass the carbon dioxide - acetone cooled trap. Carbon dioxide - acetone cooled trap to condense any substance that did not boil lower than ~80° . One—way stopcock Operated in conjunction with stOpcocks F and 3. This was closed when L was adjusted to by-pass the trap. hercury trap to prevent back pressure due to the cooling down of the system toward the end or the reaction. Safety bottle. Gas nitrometer containing 50 per cent c.P. potas- aiun hydroxide for dissolving the carbm dioxide. NMQDWXK 13 The nitromcter was of 300 ml. capacity with e dieltter or 25 an. in the vertical section and 10 mm. in the curved section. Mercury was placed in the base to seal the poteeoiun hydroxide fro- trap L. ‘ I. Nitrometcr leveling bulb consisting of e 250 I1. eeparntory tunnel. Dewar flasks used to hold the bath solution. and totnaina tain the traps G and I at their desired temperature ere not shown. III. gondenration, Separation and Fractionation Two mole: or’nethyl ethyl icopropyl carbinol acre condensed with benzene in the presence of enhydrouu aluminum.chloridc. The general procedure followed one that developed by Huston end oeverel earlier‘workorl in this laboratory. . It hoe been eotebliohed by Huston and Fried.- nnnn (1) that the proportion of benzene to alcohol should be about five moles to one mole respectively. Kuoton and Barrett (2) end Huston and Smith (3) have found that the proportion of aluminum chloride to alcohol should be anew hnlffimole to one mole respectively in order to obtain the maximun.y1eld of the expected alkylbonzene condense» tion product. Two condensation: were carried out using these proportions. A typical condeneetion will be de- scribed belou in detail. 14 Sixty-mix and eight-tenths grams (i mole) of 6.1’. anhydrous slulinun chloride was placed in the reac- tion flask with 390 g. (5 U019.) of anhydrous, thiophsns free bonsene. This mixture was heated to reflux‘with stirring to drive off large amounts or air dissolved in the beusne. Hydrogen chloride was evolved in quantity and the alusinun.chloride suspension took on a reddish color. The solution was then allowed to cool to root teln- ereture. At this time the system was swept out with (:02. “ms nitroneter was connected and the flow or gas adjusted until.sdcrobubbles appeared in the nitroseter. The ice- salt trap and dry iee~acetone trap were connected and the alcohol was added with stirring at a rate Just fast enough to maintain the temperature between 30° and 35° c. After addition or approximately one-half mole of the alcohol, considerable back pressure was noticed. This may have been due to a decrease in the gaseous products being formed and a subsiding or the heat or reaction. when the addi- tion.ues couplete, the product was stirred for one hour and then allowed to stand overnight. At this time any gas which for-ed in the nitroaeter was transferred to gas burettes and analysed. Volatile liquids round in the traps were purified by distillation, their boiling points determined and then sealed. During thece condensations only a very small amount or liquid was found in the gas traps. In an effort to determine the boiling point, a tiss-tmerature curve was run revealing a depression or the curve over an approximate ten degree range close to 15 ~23° c. and again at about -10° 0. These depressions are in the boiling point range of methyl chloride and iso— butane respectively. Theorw'would predict these compounds as being present. The small quantity or liquid made sep- aration.or these liquids and determination or physical constants an impossibility. Gas caught in the nitronetcr'was transferred to gas burettes and tested for condensibility by passing it back and forth through a tube isnersed in.a dry ice — acetone bath. Only*a very small alount condensed so that positive identification'was still not possible. The gas did, however, burn with a blue to bright yellow flame, giving additional evidence of the presence of Isthyl chloride. The reaction product was hydrolyzed by the addi- tion of cracked ice and stirring. The temperature was maintained below 35° c. until addition or a mu a-ount or ice caueed no rise in temperature. Then 100 ad. or water was added. li‘he organic layer was separated tre- ths aqueous layer and the latter extracted three tiles with 25 ml. portions of benzene. The combined extracts and organic layerbwas dried over anhydrous sodium sulfate. Distillations were carried out with a 60 on. helice packed column. The head on the column was designed for distillation under reduced pressure with a takecorf forvpennitting reflux. The condensation products were combined and the solvent distilled off at atmospheric pressure. A carbon dioxide - acetone trap was connected 16 to the head of the column at this time to condenee any low boiling liquid. dissolved in the solvent . The wt of. liquid caught in this manner wee not sufficient for further investigation. Previous nowhere in this laboratory (17) (3) have found it necessary to remove the city). chloride fraction in order to prevent contamination of the manor boiling fractions. Since Binder (6) had previously ino— lnted and identified this fraction in quantity and had difficulty in taking any further ”petition of the high boiling fractions, no attent was made to isolate thie product. It no removed by refluxing the combined frac- tions with an equal value of fifty per cent alcoholic potauinn hydroxide for a period of four bouro . The organic layer was then separated from the water layer and the former neehod with small portions of water. The organic layer was then dried over anhydrous potassium carbonate. On subsequent distillation the high boiling fraction. were found to be free of chloride. Due to the large hold-up in a 60 on. column, it nae found neoeoeary to purify each fraction using a smaller colunm of the can type. Upon two or three repeated distinctions, fractions were obtained lufficiently pure for the preparation of derivativee after the class of wound had been determined. An appreciable quantity of uteri cl boiled higher than the expected hoptyl benzene and in suspected of being 17 di- and try-substituted alkylbensenes. A snell tarry residue remained from the distillations. IV. ldggtification gt; Fractions Fractig ;: This fraction uas isolated as a liquid in the dry ice ~ acetone trap. in attempt was made to da- temine its boiling point as described in the section on condensation, separation and fractionation. ho definite boiling compounds were isolated. The mixture boiled over a range of tree ~3o° to 10° c. Since this is over the boiling range of both methyl chloride and isobutane, they are suspected of being present in this fraction. Gases condensed and isolated in this manner burned with a blue to bright yellos‘ flame, giving additional evidence of the presence of lethyl chloride. Practig g: Tertiary Butyl Benzene. Three grams of the hydrocarbon per sole or alcohol condensed was isolated. This compound was identified by means of its acetenino derivative. Physical constants of the hydrocarbon are recorded below: field: fi ”11 ' 53° c s ‘2’? 3 1 s‘926 Helting point of acetamino derivative: 168.5- 169.5° a. Physical constants checked with the literature 18 (27) and the uniting point of the derivative attacked with the work of Ipaticff and Schnerling (26) and with Barrett (2) . one method of preparation was that given by Ipatieff and dormer-ling (26). The nitro compound was prepared by treating 1-2 cc. of the alkyl benzene with 5 co. of a 1:1 mixture of concentrated nitric and sulfuric acids. when the reaction of aixture had cooled down, it was poured onto crashed ice and extracted several times with ether. the extracts were washed with water and the other evaporated. The nitro W was dissolved in a few cc. of alcohol and 5 g. of granular tin were added. About 5 cc. of concentrated ml were added dropwise with shak- ing. The mixture was shaken until a few drops tested in water showed no marked turbidity. he mixture was then allowed to stand 30 minutes to insure complete reduction. The mixture was decanted into 20 cc. of water and the tin hydrochloride complex salt of the amine extracted with other: and #05 sodiu- hydroaide was added to free the stains. in. anine was extracted with other, washed with water, and dried over anhydrous potassium carbonate. The other solution was filtered from the drying agent and the ether evaporated. The acetyl derivative was node by adding 1-2 cc. of acetic anhydride to the amino coapound. The excess anhydride was then hydrolised by warning with 5 ml. of water and the solution evaporated nearly to dryness. The impure solid was filtered off l9 and washed free of acetic acid. It was then recrystallised m 50% auMa Fraction III: Tertiary Amy]. Benzene. A total or seven grass or this compound were isolated and were identified by moans of the acetamino derivative. The following phys- ical constants were determined: Yieldi 2.455 8171.31 190° c. ”15 I 88° c. 32,? z l.#917 . lelting point a: «stain derivative: 137-138° c . The acetamino derivative was prepared in the manner previously described. i'hs melting point or the deritative checked with the work or Ipatierr and Schner- ling and with.Barrett. The physical constants of the tertiary amyl benzene checked with those found in the literature (2?). Fraction 11: 3-hethyl-3aphenylpentane. The physical constants were found to be as follows: Yield: 5 areas ”750: 215-218° a. $13 8 85-830 0. ago : 1J9h5 An attempt was made to prove the structure of this command by scans of its acetanino derivative . 20 However, the small yield of uterial and the large amount of cmtanination fro- thc heptyl benzene fraction Just above it and the unsaturated fraction Just below made purification impossible, and no satisfactory derivative was obtained. It was interesting to note that in the small fraction immediately before this, there occurred a very sharp decrease in refractive index. The decrease was free 1.169% to l.t9lo. This is a very strong indication of the presence of unsaturated aliphatic compounds. when tested with a 55% solution of Bra in 0611;, this small frac- tion gave a very positive test for maturation. The DJ. of this fraction is in the range of 0131-125, and our theory would predict the presence of both 3-hsthy1u-3ephenylpentane and these unsaturated «mounds. The mechanism of their formation will be tahsn up sore fully in the Discussion. It is felt, however, that the evidence points strongly to the presence of these W in this fraction. Fraction 1: Methyl ethyl isopropyl phenlenethane. This fraction was the expected condensation product, but a poor yield was obtained. The coupound was isolated in a very pure form, however, and the physical constants reported below are in very close agreement with those recorded in the literature (28) (23). 21 Yield: 95 31mg: no.5" c. BF”, 3 99° C. a? x 1.5012 D3? 8 .8836 20 : 31.23 dynes (Du Hon y) 29.02 dynes (Drop wt.) a carbon and hydmgen analysis or this compound has been run by Binder (28), and Rodrick (23) has pre-» pared a derivative of it by converting it to the phenol. Since all physical cmtants were in excellent agreement with the work or these investigators, it was felt that these additional checks were mosessary, and that there is no doubt as to the identity of the ccnpmmd. Fraction 31;: An attempt was made to purify this fraction, but no definite separaticm could be lads. It is suspected or being a mixture of di- and tri-substituted alkyl hen- tones. It distilled above 123° 0. o 3 ass. and over a very wide range. Yield: 31 grass Fraction 111;: Terry residue Yield: 5 grams 22 A3 7.5- wnt «553. 4.5.5.3 am 05:80 13%.: 1.me «Sm; mamm H3233 "Pug ”have: .mp3 «.5» on: 33ch 38353 «3 m3.” mama.” @863 AacefiééhfiunA mm” .. Rn A... Em. A...- 32 h .mfloo 23833.“. 25.” :34 cm” 3833 up: 5.8339 om.m3 .. 9mm” .9: :3 TE. amt m 528 ofiaouooaun m3»; mmmaé mm: 8233 H33 gong 333 30% 5333353 no .93 33 «acid 33353 833:. 233. .3 9'32;th cam: H Mega 23 DISCUSSIOH In condensing methyl ethyl is0propyl carbincl with benzene, in the presence of anhydrous aluminum.chlcr- ide, Binder (6) found that the yield of the expected hep- tyl benzene condensation product was very low. This was apparently due to considerable fragmentation and rearrange- ment or the alcohol during the course or the reaction. The purpose or this investigation was to separate and identify these products of fragmentation and rearrangement. It has been shown by previous investigators in this laboratory (2) (3) (17) that fragmentation takes place to a much greater extent with the highly branched alcohols than with the non-branched alcohols. The large amount of fragmentation found in this study in consistent with these facts. Uhitnore (22) has found that aluminum chloride can cause molecules to rearrange by methyl group aigration. Bcedtker and Halse (28) have observed a rupture of the molecule with aluminum chloride as catalyst. Both rear- rangement and chain rupture are evident in the study of this alcohol. At the present time, the host generally accepted theory for the condensation mechanism is that or Price (19), as nentioncd in the theoretical part or this thesis. On the basic of this theory, the carboniun ion.may be formed from the alcohol, the chloride, the alkene or the alkyl benzene. The electron deficient cation then receives its 236 pair from the double bond of the benzene ring. The form- ation of the condensation product proceeds in the following manner: H a; . gag (1) (2113:?- on + A1c13 ————, c113 + + Al(m)013" CH3 8 CH3 1! ¢H3 33 3 3 an 2 a 3a. a e . WC!” ++n n-—»033 '. n-bcn33‘za ’3 an, an 3" CH3 H CH3 The fcmtion of the carboniun ion in equation (1) say now lead to a number of possible side reactions. Cos- bination sith the chloride ion present in large quantities would occur as in equation ( 3). This compound has prev- iously been isolated from the condensation of this alcohol by Binder (6) and no attemt was made to isolate it in this study. W3 3 the (3) 633.3; + Cl” 3' (2:3 - 01 633 3 33 Loss of a proton from the carboniun ion would lead to the formation of an olefin. 25 33 Ha CH3 2113 (’3) 0023 1- (loss of proton); CH3-¢ .. : CH~CB3 3 (‘33 and OH H cn3-t g S 3-032-033 Neither of the two alkenes predicted by equation (‘5) acre isolated in this condensation. however, the solvent rs~ moved frm the condensation products Save a positive test for unsaturation on treatment with a 5% solution of Dr; in 661;. The boiling point and refractive index of the solvent were also significantly lowered. On treatment with bro-inc the benzene was separated from the contami- nant but the residue was not large enough to enable a satisfactory separation of the resulting mixture. Since the boiling points of these two alkenes are 86° and 97° 0. respectively, it is quite feasible that they may have been present in small amounts. The formation of t-butyl benzene and t-anyl bensene and the resulting low yield of the condensation product, may be attributed to the tendency of the cation to undergo rearrangement sith a resulting rupture of the chain. A e 5 cs " - cs-cs ( ) 6113- ca3 " 3 Proton llhlfl anaemia 26 The t-amyl carboniun ion immediately hooks up with the benzene nucleus to tors the alkyl benzene. In a similar manner so can account for the formation or t-butyl benzene. 3 , 32 3-91: (6) 683': 8M1$w3§28hiwtm§3fi “318?? 033'- “3 + anyone-cu Proton Shin cam The t-butyl cation may then react with the benzene nucleus to give the alkyl benzene or a proton shift say tales place to tom isobutene. It appears from previous work in this laboratory ( 3) (5) that isobutenc say condense with benzene or be immediately reduced to iscbutane although the Incoh- anion of this reduction is not yet clear. This possibility would account for the isobutane suspected of being present in the gas traps. I: the cation in equation (5) were to Imdergo dmthylatien during the methyl shirt , we can acwoount for the formation or such products as nethyl chloride and 3-Iethyl-3-phenyl pehtane, both or which are strongly indicated as being present among the condensation products. Evidence of such denothylation has been presented by Van Byte (17) and Barrett (2) in earlier investigations in this laboratory. The mechanism might be as follows: 27 H3 3 + 315% (7) 033- + (demotrqlationl; CH3~§E CH3 , an - ' 3 "_'~ 3 am, so CH3-Cfiz-gn; CR~OH3 (8) mg}: + c113 ————. 03301 + H“ rho tortiarw’hoxyl cation for-ed could than hook up with a benzene unolouo to foul tho nlkyl tenuous. It 1. quit. possiblo that u.dolnthwintod hoptyi cation, tonlod 1n oquntion (7) night attack mother hoptyl cation to follow" 3 3 (9) on??? g: + :gagaé 03.033 -—+ 013-??- E; gag-g”?- cue-CH3 u Proton Shirt 1. CH Olly-EB; dug - § - $3; 333 8” an: CH3 Such a 319on u. the one postulated above would account for the formation or an unsaturated 6131126 motion such a :- thought to be present in motion :11. Evidence of the formation or such compounds ha been presented by Kraut: (5) in an earlier investigation: in thin laboratory. 1. 2. 3. ll. 28 SW? lethyl ethyl ioopropyl carbinol has condensed with bon- Icno in the promo or anhydrou- aluninun chloride. Only 9% of tho expected methyl ethyl iaopropyl phonyl nothano In ominod. Tertiary botyl bench. and tertiary anyl beacon. home been isolated and their structural proton by noon- of their acotamino derivativu. 8mm othcr mammary product: our. incl-tad but their otruoturou were not definitely proton. loohcnim have boon cogent-d for tho {oration of all products. 1. 2. 3. t. S. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 29 BIBLIOGRAPHY Bhutan and Friedemann, J. Am. Chem. 800., §§, 2527 (1915) Hutton and Barrett, 11.8. Thesis, Mich. State Coll.(l9t2) Huston and Smith, 14.3. Thesis, Mich. State 00114191.?) Huston end Awuepere, ibid., (19%) Huston and Irentc, Ph.D. Thecicmmh. State Coll'.(191l7) Hutton and Binder, M.S. Thesis. Hich- State 0011-(1935) lief Ann., 228,. 255 (1892) Huston and Sager, J. Am. Chem. 800., ‘fig, 1955 (1926) Huston and Davie, “.3. Thesis, Mich. State 0011.(1933) Huston and Goodemoot, H.S. Thesis, Mich. State Coil.(193#) Tzukervanik, J. Gen. Chem.(U.S.S.R.), 5. 117. 764. 767. (1935) Huston ma 11mm, .7. Am. Chem. 800., g, #39, (1936) Huston, Fox, and Binder J. Org. Chem, 3, 251. (1933) Huston, Guile, Schulati, and Hanson, ibid., g, 252 (19131) much and Hughes, mm. T115513 ,mch. State 0011419100) Huston end Kaye, ibid., (1942) Huston and Eaterdahl, 31.8. Thalia, Mich. State Coll.(19t0) Hutton and Curtis, 1b1d., (1951) Welsh and Drake, J. Am. Chem. 800.. ‘QQ. 59 (1938) Norris and Ingrahan, 11.14., _6_g, 11.21, (1938) lorrio and Sturgic, ibid., Q, 1H3, (1939) Huston and van Dyke, Ph.D. Thecic,Hioh. State Coll.(l9#§) Roxanna and Sewn, J. Am. Chem. Soc. 22, #70 (1937) Price, Chem. new, 2 , M (19“) Uertypmch and Firle, Ann., 599, 287 (1933) 21. 22. 23. 24. 25. 26. 27. 28. 29. 30 Ulioh end Reyna, z. Electrochem., 3;, 509 (1935) Whitmre, J. .1121. Chem. Soe.,§_4_,327h (1932) Huston and Hedrick, Ph.D. Thesiemzich. State Coll.(1937) Gilman, Wilkeneon, Pinchel, and Meyers, J. Am. Chem. 800., 32‘, 150 (1923) Whitman and Evert, J. Am. Chem. 300., 25,82 (1933) Ipatieff and Schmerling,ibid., fig, 11.78 (1938) 1mm, ibid., 52, 1056 (1937) Eglorr, ”Physical Constanta or Hydro- carbons," Vol. III, Reinhold Publishing Corp., New York, 11.2. Huston, Fox and Binder, J. Org. Chem, 3., 251 (1938) Boedtlcer and Heine, Bull. Soc. cum” 12, w. (.1916) M7|)fifl)“\i|flifl)@\(“Eijfflflififlljfiflmli'ES