Mill IHMHIHIHI l l 1 WW . 145 326 THS A STUDY 0-? WE S‘E’EREC WCTS IN THE REAfiTfON OF AWL ESERS Wi‘EH AN ARYL GREGNARD REAGENE Thesis for ”we Degree of M. S. MICHIGAN STATE UNEVERSITY Robert Eugene McComb 1957 MICHIGAN STAIE UNIVERSITY EAST LANSING, MICHIGAN LIBRARY Michigan State University ‘t' “u" '7’ “L”; .'. 3‘. "m ...-. an 3‘9wa, ‘9’ If“ m 13 2-rr::.I3‘m 1:4 mm... . ,‘i _‘(' .......H:' r," ‘1‘: '. ,. .“ -,..,_ , .. (x V" _1__ 0‘ ‘~“vTIvh t? AnYu maiawu filld AN ner Unlunanu finAGENT 3? Robert Eugene ficComb A THEBIS Submitted to the College of Science and Arts of michlgan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of QASTER OF SCIENCE Department of Chemistry 1957 I m. 7‘7 I. ”:l I? i A, ‘- 6_ .. j 3: tr . ~, ./ ACKNOWLEDGEERT The writer wishes to exoreaa his appreciation to Doctor W. T. Lippencott for his guidance and assistance through- out thin investigation. fiffififlflfiflifiifififl 11 A STUDY OF THE STZRIC EFFECTS IN THE REACTION OF ARYL ESTSfiS WITfi AN ARXL GEIGfiAfiD BEAGSHT BY Robert Eugene KoComb AN ABSTRACT Submitted.tc the College of Science and Arts of Eichigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry Year 1957 Approved Kk/uf/f/A 444A? 0-.- ..~- ABSTRACT A study or the steric effects of ortho substitution on eryl esters in the reaction of these esters with meeityl- magnesium bromide has been made. fleeitylmagnesiun bromide was caused to react with the phenyl end'g-oresyl esters of benzoio sold, and gymethyl-,‘2-methyl-,‘gyethylo, and‘gfiso- propylbenaoic acids. The reaction mixtures were analysed quantitatively by infrared absorption spectroscopy for the diaryl ketone and triaryl tertiary alcohole formed and for the unreacted aryl esters remaining after the reaction. Verification of the infrared method was done by determining the percent ketone formed as an oxime using hydroxylamine hydrochloride in pyridine and percent unreacted ester by hydrolysing the ester with excess bees and back titrating the excess base. Results of this study indicate that ortho substitution of alkyl groups on either the eroyl or aryloxy portion of the ester hinder the reaction of these esters with the highll hindered Grignard reagent. 0n the other hand, para substitution of the methyl group enhanced the reactivity of the esters with the Grignard reagent. The reactivity of ester with uesitylmagnesiun bromide was virtually independ- ent of the size of gpslkyl eubstituent on the aroyl group. it The greatest decrease in reactivity was found for ortho sub~ stitution on the aryloxy portion of the ester. The reactivity of the complex formed was the reverse of the reactivity of the esters. The ortho substitution on the eryloxy portion of the ester increased the reactivity of the complex when compared to unsubstituted esters. The ortho substitution on the sroyl portion likewise increased the reactivity of the complex when compared to unsubstituted esters but this increase was not as large as for substitu- tion on the sryloxy portion of the ester. The size of the ortho substituent on the sroyl portion did not greatly effect this reactivity. TABLE OF CONTfifiT3 INi‘RODUCTIC;N.OOOOOOOOIOOOOOIOOOIOOIOOOOOOOOOOOOOOOUCOO Statement of Problem........ ....... ................ luatorioalOOOOOOIO0.0IOOOOIOOOO0.00...0.0.0.0090... EXPEF‘vIlfAISN‘I‘ALOOIIOI...eoeeeeeeeeeeeseeeeeDoses-00.0.0.0 sources and Preparation of Starting ssterials...... Preparation of gofithyl- and g-IsOprOpylbenzoic Acidoeeeooooeeoeeeeeeeeeeeeesseseeeeeeeeseeeeeeeeee Preparation of Acid Chlorisee...................... Preparation of Esters........... ................... Preparation of Grignard Beegents................... REI‘CTICN 0F GRI(}PE.AFD FLdJi-J’;;‘w'j‘$ VIIT-i 231.1323 sees-esseeeee. AKEALYSIS CF! Ifio’; WU Ts..oeeOeooeeOIeeeoeloecoo-0.90.0000 Determination of Ketones by Infrared bethod........ Determination of Alcohols by Infrared hethod....... Determination of Unrsscted Esters by Infrared EcthOdOOOOOOOIOOCOOIQOIOOOOOIOOCOOOCOOOOIOOCOO.I.0. RESULTS-oolcooeoeoeese9seenso...eeeoeoeoleneseooosoee. OCOIOOOOOUOO DISCUSSIONOOOOO.OOOOOOIOOO...0.0.0.0000... CCE‘ZCLUSIObiIOOOOOOOIICOO...OIIOOIOODIOOOOUOOOOIOOO {'35 IBLI'mRAPgfiOOOOO...0.0.0...90.000.00.000.IOIOOOOOOOIO AP FEr‘fL‘IXIOOIOOO.OOOIOOIOOOOOOOOOOIOOOO0.00.00.00.00... vi IV. VI. VII. VIII. IX. XI. xlle LIST OF TABLES Page Yields and Physical Constants of Esters ..... ‘ PreparedOOOOOOODDOOOOOIiOOOOOOOOOOOOOOOOOOOOOO 16 Initial Concentrations of Reagents in Fee. actions of Hesitylnagnesiun Bromide with Aryl Estera'OOOO000.000.0000...ICQOOOIQIOIOICI 20 Infrared Determination of Diaryl Ketones, Triaryl Tertiary Alcohols and Unreacted Aryl Esters in Reactions of hesityl- . magnesium Bromide with Esters................. 23 Determination of Diaryl xetonee by Hydrosyl- amine Hydrochloride and Unreacted Aryl Esters by Hydrolysis.......................... 24 Infrared Transmittancy for Standard . Solutions of Bensophenone in Toluene.......... 26 Results of Infrared Determinations of Ketones in Reaction Eixtures.................. 28 Infrared Transmittancy for Standard Solutions of Triphenyl Carbinol in Toluene.... 29 Results of Infrared Determinations of Alcohols in Reaction Eixtures................. (A to Infrared Transmittancy for Standard Solutions of Phenyl Bensoate in Toluene....... 33 Results of Infrared Determinations of the Unreaoted Esters in the Reaction hixturee..... 85 Percent Ester Reacting, Percent Ketone Complex Formed, Percent Ketone Complex Con- verted to Alcohol and Percent Grignard Reacting in the Reactions of Eeaitylmag- “0.1“ Emmldg 'ith tho Ari]. E‘teraeeeeeeeeaee 41 The Reactions of Ester, Ketone Complex and Grignard, fielative to Phenyl Bensoate......... 47 711 2. 3. 4-8. LIST OF FIGUR€3 Page Standard Curve of the Ketone.................. 27 Standard Curve of the Alcohol....... .......... 30 Standard Curve of the Ester...;............e.. 54 Representative Infrered Absorption Spec- trograms of Reaction mixtures of 9T?§??T?..... 65-60 Reagent with Esters..... ..... viii INTRODUCTION 22335532310? 32132919.. _ . s ., ,_ _ The effects of steric hindrance of alkyl groups in the reaction between Grignard reagents and esters or ketones have been widely studied (1-15). But until recently (16) no systematic study of the interplay of sterio factors in the reacnion of Grignerd reagents with esters has been reported. There are three sterio areas involved in this reaction. These areas are the acyl portion of the ester, the alcohol portion of the ester and the Grignard reagent. This investigation is specifically a study of the effects of steric hindrance on the reaction of nesitylmagnesium bromide with the phenyl and 2-orssyl esters of benzoic acid and.g-methylo,.g-methyl-,‘2-ethyl- and g-isOprOpylbenzoic acid. The 2-methylbenzoates were used to measure the elec- tron donating effect of the alkyl group. The purpose of this investigation was to determine the nature of the interplay of the steric factors of the esters when the steric hindrance in the Grignard reagent was ex- tremely high. It was also desired to determine how sensio tive the reaction of the formation of alcohol was to sterio factors. Finally, from this investigation it was hoped that more information would be obtained on the mechanism of this reaction. The study was carried out by causing a 1:1 mole ratio of Grignard reagent to ester to react in refluxing toluene ' for one hour. Due to the difficulty in initiating the forma« tion of Grignerd reagent of bromonesitylene one drop of ethyl bromide was employed to start the reaction. The Orignsrd reagent was added to the ester which is the reverse of the normal addition so as not to favor the formation of tertiary alcohol. The Grignard reagent was filtered before use to remove any excess magnesium so as to step any unde- sirable side reactions. The products, ketones and tertiary alcohol and the unreected ester were determined by quantita- tive infrared absorption methods. The infrared method was checked by e volumetric determination of the axles deriva- tive of the ketone and by ssponificetion of the unreacted ester. Afiistorical The reaction betseen Grignard reagents and esters and the effects of steric hindrance in both the Grignsrd and the ester have been studied by many investigators. Normal addi- tion, formation of tertiary alcohol, reactions which lead to formation of hydrocarbons, reduction, enolization and other reactions have been included in these studies. Boyd and Hott (1) studied the reaction between 2-methyl- phenylmsgnesium bromide eith ethyl g-toluate and found that no tr1-2-tolyl carblnol was formed. They isolated a disryl . 9 , "4 o _ ‘A a n .. . - I . . >.. . r ‘. . o t I Q , . .. ‘ a u U . 1 . p . - . . l o . . ,. .. . . a v ketone and the pinecol of the diaryl ketcne and postulated that thio formation was due to the excess magnesium present. Kohler and Beltzly (2) studied the reaction of aceto- mecitylene, dimeeityl ketone and dimeeityl diketone with ethylnagneeium bromide. They found that the eteric hindrance due to the methyl substitution in the ortho position stopped the normal reaction of addition to the carbonyl and caused enolization in the case of the acetomeeitylene, but for the other two compounds no reaction could be detected. They ex- preeeed the thought that the decreased velocity of the re? action, because of sterio factors, allowed other slower re- actions to take place. Complete blocking of the functional group stopped any reaction from taking place. Smith and Guee (3) in etudiee similar to thoee of Kohler 1nd Bultxly verified their results. The elkyl ketonee hin- dered near the functional group gave methane when reacted vith methylmegneeium iodide. This was apparently due to enolizetion of the ketone. This enolizetion was a function of the hindrance in the ketone, and the greatest effect was found when the alpha position one hindered by methyl groups. Arnold and co-workere (4) investigated reactions be— tween phenylmagneeium bromide and ellyl benzoetec, tolueteo and ieoduryletea which were substituted in the ortho posi- tion. They found upon reacting phenylmegneeium bromide With allyl benzoate and cllyllg-toluete yields of 86% and '7‘? J. \-.A 68% respectively, eere found of the triphenyl carbinol. They felt that one gynethyl group did not greatly hinder the normal addition of the Grignerd reagent. When they reected phenylmagneeiun bromide with ellyl [9 ~1eodurylete, they ob— tained ellylbeneene, e negneeiun heloeerboxylete end no alcohol. Thin wee interpreted as eteric hindrance etopping the normal addition. They investigated also the reaction between ellyl trimethyleoetete end phenylmegneeiun bromide and found the normal addition to take place which gave tert- butyldiphenyl carbinol. '_ Pueon end co-workere (5) etudied the reactions or certain alkyl and eryl eetere of neeitoic eoid with elkyl end eryl Grignard reagents. The eXpeeted products eere not found even though e reaction wee in evidence. with the elkyl mesitoetee end elkylmegneeium helide, they found meeitoic acid and en elkyl halide which came from the cityl group of the ester and no elkyletion eee observed in any of these re- actions. With the eryl meeitoetee end elkylmegneeiun helidee they cbteined elkyl neeityl'ketone and phenol. when both reeotente were eryl compounds, the: found eryletion or eub- etituticn in the ortho position of the ketone produced. Whitmore and George (6) reacted different Grignerd re— acente with dl-llOpPOpyl ketone and round that enolizetion increaeed from O to 90% going from methylmagnesiun bromide, ethylmegneeiun bromide, eec-butylmagneeium bromide, tert—butyl 9p. magneeium bromide to neopentylmagneeium bromide. This enolization one due in part to the steric hindrance and caused a decrease in the reactivity of normal addition. Whitmore postulates a cyclic intermediate to account for thie. Haueer and co-Iorkerc (7) studied the reactions of car- boxylic eaters eith different Grignerd reagents. They found that en the Grignard reagent became more complex (from methyl, ethyl, ieopropy1,.£§££-butyl to mesitylmagneeiun bromide) the normal addition dropped off. Under these con- ditions condensation reactions and some enolization occurred. Trieb (8) studied the effects of aubetituente in esters on the reaction rates with methylmagnesium iodide and he postulated the following rules for the eater RCUOE‘: 1. Increase in the size of normal R' up to prepyl causes decrease in reactivity, negligible thereafter (sterio). 2.. Benzyl esters are the moat reactive, but fur- ther separation of the phenyl by a normal car- bon chain decreases the activity (polar). 3. Branching in R' inhibits when adjacent to 000- but accelerates when removed by at least one carbon atom (polar and sterio). 4. Diaetereomeric eeterc react at different rates I}: (isccentnyl> neoisonenthyl) nenthyl) nec- Ienthyl acetates) (sterio). 5. Phenyl groups in R cause acceleration, but the effect is dicinished by separation from ~coo (polar). 6. Branching in R causes retardation in rate (sterio). Whitcors and Lewis (9) and other cc-workers (10-14) studied the effects of substitutions on the ecyl port of alkyl esters, elkyl acid chlorides and ketones. When substi- tution was on the alpha carbon, they found that an ethyl group produces a more profound effect then a methyl group and that as the size of the alkyl substituent increased the normal addition reaction decreased in importance.” This veri~ fied Trieb's sixth rule. In discussing sterio hindrance in the reaction of Grignerd reagents with esters, Kherecch and Reinnuth (15) commented that Trieb's (8) studies showed that substitution on both the acidic and alcoholic constituents influence the addition rates. It was established that these influenced were more sterio then energetic and that the abnormalities in the reactions observed were due solely to these sterio factors. Therefore, branching in the acid chain of an alkyl acid near the functional group retarded addition of the Grignerd reagent. Branching in the alcoholic portion of the A...» ester had less influence etericelly on addition but the effects were still noticeable. They suggested that the alco- holic portions influence was more of an energetic effect than sterio. These postulates and suggestions were based on a wide variety of experimental results conducted under many different conditions. Kharsech and Reinmuth also pointed out that if the intermediate first formed was ether soluble, the ability to form ketones was a function of the Grignerd reagent and not the ester. 'In this view it was shown that branched alkyl4 magnesium halides and diortho substituted arylmegneeium halides form a ketone preferentially. This would suggest that the effect is sterio rather than energetic. If the intermediate was ether insoluble, the formation of the ketone found was primarily due to the ester and not the Grignard re- agent. The literature referred to no systematic study of the effects of ortho substitution of methyl, ethyl or isOpropyl groups on the aroyl portion or of gymethyl substitution on the aryloxy portion of the ester. Until recently (16) this study seemed to have been ignored, though it apparently has been done partially for ketones. EXPERIMENTAL The following esters were caused to react with mesityl- assassins bromide: phenyl bensoste, phenyl‘grmethylbenzoste, phenleQ-sethyibensosts, phenyi‘grethylbenzosts, phenyl ‘ggisopropylbensoste: grcresyl bensosts,‘ggcresyl.g«msthyl- benzoste,‘gycresyl‘grmethylbenzoste,‘g-creeyi‘g-ethylbenzo- ete,.g-cresyl.g-isopropylbensoste. The resulting reaction mixture was analysed by infrsred snelysis for the smounts of ketons and tertiary alcohol produced end for the smount of ester that remained unreseted. Further snslyses were per- formed by determining the ketones in certsin of the rescmion mixtures using hydroxylsmine hydrochloride in pyridine, and the unrsscted esters by hydrolysis, with en excess of stand- ard bees in alcohol and back titreting the excess base with stsndsrd scid. SOURCEQWAhD FREPARATIOH O? STdRTIfiG ghEERIALS The chemicals used in this study were: phenol,'g-cresol, benzcyl ohlorids,‘g¢toluic acid, Eytoluic scid, thionyl chloride,.g-bromoethylbensene,‘grbromoiseprOpylbensene, and mssitylene. Compounds prepared from the above were phenyl snd‘gecreeyl benzoste,.grmethylbenzoyl chloride, phenyl and ‘groresyl,g-methylbensoste,‘grmethylbensoyl chloride, phenyl and.g-cresy1,g-nethylbenscste,‘gyethylbsnsoic scid,,g-ethy1- bensoyl chloride, phenyl and g—cresyl,g-ethylbenzoste, [3-iaoprcpylbensoio acid,.g-isopr0pylbenzoyl chloride, phenyl and‘g~crcayl‘geincpropylbenzcate, and bromomeaitylene. rateriala Purchaeeg Phenol, Eerck and Company, Analytical Reagent Grade. .g-Bromosthylbenzene, The Dow Chemical Company, radian tilled by Arthur J. Pastor and traction boiling 95°C at 30 mm. {3. was used, n925 1.5452. ,g-Bromoiaopropylbenzene, The Dow Chemical Company, re- dietillod by Arthur J. Pastor and traction boiling 107°C at 30 mm. Hg. was used, “926 1.5585. ,g-Toluic acid, Eaitman Kodak Company, white label No. 1646, m.p. 179-18006. acuityleno, fiathaaon Coleman and Ball Eiviaion, white label, b.p. 153-15500, n935 1.4987. Benzoyl chloride, J. T. Baker Chemical Company, b.p. 195.1-197.1°c, 99.9; purity. Thionyl chloride, Bantman Kodak Company, white label. ,g-Creaol, Eastman Kodak Company, yellow label (Technical), it was distilled twice through a cropped 40 cm. Vigreux column and traction boiling at 93-94°C at 25 mm. Hg. was used. wagnenium metal turnings, Merck and Company, specific for Crignard reactionl, 99.5} purity. Pyridine, Eerck and Company, Analytical Reagent Grade. 10 Chlorctorm, Rerck and Company, U.3.P. Ethyl alcohol, Union Carbide and Carbon 00., "03". 95/50 Preparation of Broeomesitxlsne Bromomesitylene was prepared from bromine and mesitylene by the method described in Organic Synthesis (1?). In a one liter three neck round-bottom flask equipped with a true bore stirrer, a 500 m1. dropping funnel and a parallel side arm holding a thermometer and reflux condenser, were placed 321.2 g. (2.676 moles) of meeitylene and 150 ml. or carbon tetrachloride. A rubber tube from the too of the condenser to the water drain was used to remove the hydrogen bromide vapors formed. The flask was cooled to 2°C and 483 g. (161 ml., 3 moles) of bromine in 190 ml. of carbon tetrachloride were added over e period of two and oneohalr hours while the tenpereturo was maintained at 6°C by means or an ice both. after all the bromine was added, the mixture was refluxed for one hour. The reaction mixture was washed with two 200 ml. portions of distilled water and two 250 ml. portions or a 55 sodium hydroxide solution. The reaction mixture wee dried over anhydrous calcium chloride and the carbon tetra- chloride distilled at atmospheric pressure through a eo Cm. Fenske column. The residue of the distillation was then added to one liter of 95% ethyl alcohol which had previously 11 reacted with 50 g. of Iodiun netcl. The reaction littnrc can then refluxed for tvo hours and allowed to ctend over night; It vac then diluted with three literc of dictilled water and acparatcd. the water layer can ruched with three 100 I1. portionl of carbon tetrachloride. The extractl were combined with the crude product end this mixture dried over anhydrouc calcium chloride. After drying, the carbon tetra- chloride wa- rcmoved by dictilling through c 40 on. Feneke column. The product can then cintillec through the same column at reduced pro-cure. The fraction boiling ct 97-101°O at 13 In. of Hg. vac used. The yield was 396.6 g.(l.985 melee) 741 of theoretical based on lecitylene, n925 1.8488- 1.6500. Prepargtiogficgvc-Ethzlben:oic Acid and o-IeOErOlebenzoic Acid ‘ These two acide were prepared by cousins the Grignerd reagents of oubromocthylbenzene and c-bromoieopropylbenzene to react with solid carbon dioxide (-18), In a 1000 :1. three neon round-bottom flask, equipped with a true bore atirrer, aboo ml. drOpping funnel and c reflux condenser, both funnel and condenser guarded by a dry- ing tube containing unhydrcuc calcium chloride, was placed lc.t g. (0.694 g.~atcmc) or 11.8 g. (0.485 g.-ctonc) or magnesium turningl end 76 ml. of sodium dried diethyl ether. Six gran. origobronoethylbenzene or grbrcmoieopropylbcnacne dL, 12 were added directly to the flask. The reaction nae initiated by crushing the magnesium with the stirrer. After the re- action had started the remaining dried diethyl ether 325 ml.) and‘g-bromoethylbenzene (68.7 3. total 74.? 3., 0.404 moles) or 2-bromoiaOpropylbenzene (72.4 3., total 78.4 g., 0.393 moles) were mixed and added dropwiee over a period of one hour. the eyltem was refluxed for another hour. The reaction mixture was cooled to room temperature and transferred eIOWIy with dried nitrogen onto 1000 g. of finely divided solid carbon dioxide. The mixture was stirred continuously during the addition of the Grignerd reagent and was allowed to stand until All of the unreacted solid carbon dioxide had eublimed. Three hundred fifty milliliters of n 10% hydrochloric acid solution and loo'g. or ice were added and the mixture was stirred vigorously until all cloudiness had disappeared. It we. then transferred to a one liter separatory tunnel and the enter layer diecarded.' the organic layer was extracted twice with 200 ml. of n 10% sodium hy- droxide solution and the other layer discarded. The basic solution was returned to the eeparatory funnel and acidified with 100 ml. of a 36% hydrochloric acid solu- tion. The acidified solution was extracted twice with 250 ml. of chloroform. The chloroform extract was washed With 200 ml. of e 1.5} hydrochloric acid solution and trans- ferred to a tared container from which the chloroform wee evaporated. 15 The yield of the crude g-ethylbenecic acid melting at ee-ee°c (literature velue, 66°C) (19) wee co 3., (0.355 melee), 82.61 of theoretical: or g-ieopropylbenecic ecid (it wee a dark ember liquid initially but cryetalized on standing) melting et eo-ee°c (16) was 43.4 3., (0.294 moles) 75.0% of theoretical. Agreparetion of Acidwphlorideg The ecyl halidee of gémethylg grmethyl-, grethyl- and griBOprOpylbenzoic acid were prepared by causing each of the acids to react eith thionyl chloride (20). In a 500 nl. one neck round-bottom flask, equipped with a Clniaen distillation head which was fitted with a ther— mometer, cork and reflux condeneer guarded by e drying tube containing anhydroue calcium chloride, wee placed 51.? g. (0.581 melee) of grtoluic acid and 63.6 g. (0.634 moles) of thionyl chloride, or 68.0 g. (0.e26 moles) of getoluic acid and 71.4 g. (0.605 nolee) of thionyl chloride, or 45.3 g. (0.301 melee) of‘g-ethylbenzoic acid end 57.2 g. (0.480 moles) of thionyl chloride, or 47.2 g. (0.286 moles) of ‘g-ieOpropylbentoic ecid and 54.6 g. (0.460 melee) of thionyl chloride. The epperetue nae tilted until the eide arm of the Cleieen heed wee almost upright end the reaction mixture refluxed gently for 6.5 houre, after which it we: allowed to stand overnight. The condenser use then removed end the apparatue rearranged to permit normal distillation. The 7‘1 - I I, l . e A ‘ o . .u ~- I F 14 unreected thionyl chloride wee distilled at atmospheric pressure, over the temyerature range 75~85°G. The system was then subjected to 50 mm. Hg pressure and a second frac- tion was collected and discarded. The third fraction, which was the product, gave 52.9 g. (0.342 moles) of cytolyl chloride, b.p. 96-10000 at 16 mm., yield 901 of theoretical; 60.8 g. (0.394 melee) of 2~tolyl chloride, b.p. 106-10800 at 20 mm., yield 92% of theoretical; 50.1 g. (0.296 melee) of g-ethylbenzoyl chloride, b.p. 120- 121°c at 30 mm., yield 98% of theoretical or $8.0 g. (0.263 moles) of g—icoPropylbenzoyl chloride, b.p. 28°C at 50 mm., yield 92% or theoretical. firepargtion_ef Phenyl:gnd o~Creeyl Feterg Phenyl endiggcresyl esters were prepared from benzoyl chloride, and from‘g-methylo,‘Eemethylo,‘g-ethyl-, and 2-ieo prepylbenzoyl chloridee by mixing phenol end.gfcresol dic- eolved in pyridine with each of the acyl halides (21). A detailed account of the preparation of phenyl‘g-mcthyl- benzoete is presented below: In a 500 ml. three neck round-bottom flask equipped with true bore etirrer and reflux condenser guarded by a drying tube containing enhydroua calcium chloride was placed 24.: g. (0.258 melee) phenol dieaolved in 100 m1. of pyridine. The fleck cooled to 5°C in an ice bath. After cooling, 28.8 g. (0.186 melee) grtciyl chloride was added with vigorous l5 agitation, the reaction mixture was stirred for three hours and allowed to stand. After upfroximetely 16 hours, 200 ml. of chloroform we! added and the mixture extracted six times with equivalent volume of 3; sodium carbonate solution and two times with equivalent volume of distilled water. The chloroform-ester mixture was dried by pouring it through three batches of anhydrous sodium sulfate. The chloroform was evaporated. The yield of phenyl‘g-metnylbenzoete was 35.8 g. (0.159 moles) 85.7% of theoretical, refractive index of light yellow liquid n925 1.5715. All other esters were prepared using nearly identical conditions. Solid esters were recrystallized from ethyl el- conol. Yield: and physical constants of these esters are given in Table 1. Preparation offlteeitylmegnesium Bromide nesitylmegneeium bromide was prepared by causing the bromomeeitylene to reect with the magnesium turninge in sodium dried ether. The reaction was initiated with ethyl bromide. In a 500 m1. three neck flask, equipped with true bore stirrer, 250 ml. dropping funnel and reflux condenser, both funnel and condenser guarded by drying tube of anhydrous calcium chloride, was placed 2.68 g. (0.11 g.-stome) of mag- nesium turningl and 10 ml. of sodium dried dietnyl ether, 4‘ 10 . . . 0vdh Guiancondhaohno-«Io 83 a 323?» n3 * intonm. . . . 98 382315...“ I 83 H sunkfiéi n 13.5... 0.30%.? no.2. ongucznzuoltfli Ah-uuoao . . ouqouccnahnuvlooi 33 a 3 .3 128...... 8004 no.3 3328» «ho-nonm- .. .13 Anna“ in do cacoucondhaouao: 1.0... or: a 3}. :55 . . oiaééa _ no.3 33:09:50.“ 38 A BC :5...— o .2...“ .2. 10.8 33.32253 3cm Huang 0.2.0 . a up. on 3 do» :2: 23 out“ . a»: «.2398 no.3 3-233 inofl amn- nnocu no :33 no «39m ‘ "83335 ace-u 32......“ 9:38 «.53.... u 33» II?» } if 4'“ i I; f Ll 4111!“ nmzimumm nmuflnm ho nkmflfimnoo A<0Hmnmm and andaHH H fl4n48 Ll . 17 6 g. of bromomesitylene was added and one drop of ethy1~ bromide. The reaction wan initiated by heating flask with a micro burner. A mantle was then used to heat the flask to keep the ether at gentle reflux. The remainder of the brooo- meeitylene (15.91 3., total 19.91 g., 0.1 mole) was mixed with 75 ml. dried ether and adfied drapwiae by means of the dropping funnel over a period of onenhalr hour. After the bromomesitylene was added, the mixture was refluxed for 4 hours with constant stirring. Fifty milli- liters of dried ether was added to bring the volume to approximately 140 ml. The mixture was allowed to cool to room temperature and filtered through glass wool into a driea bottle. To analyze the Grignard reagent, two milliliters of it was pipetted into 10 ml. of distilled water and a known excess of 0.6165 M hydrochloric acid was adéed. The excess standard acid was titrated to a Methyl Red endpoint with 0.6163 N sodium hydroxide solution to determine the number of milliequivalence per milliliter and rough yield of Crignard reagent. The method of calculating the milliequivalenta per milliliter and percent Grignard reagent is given below: millieq /ml : E1. of 901? x N 8911 = r '. c 2.46 m1- 3 06165 N . case millieq./m1o r» 18 g Grignard raasent .‘gillieqL/ml. x ml._pf Grisnarg*x#lfio a #fTEU miIlieq. 0.758 x IFO a 91;. The remaining portion of the Grignard reagent was used in the reactions with the esters. ‘figaationo of4heqigfilgzgg§§ium Brcmide *1 h Ea Esta L0 Approximately 0.02 mole: of mesitylmagneaium bromide was mixed with Aparoximately 0.02 moles of each of the ester: and allowed to reflux for one hour in 100 ml. of toluene. The reaction mixture was then quenched with aqueous hyiroohlorio acid and worked up for analysis. An example of a typical re- action is given below: In a 250 ml. three neck round~bottom flask equipped with a truc bore stirrer, Dean-Stark take-0ft tabs fittea with a reflux condenser guarded by a drying tube containing anhydrous calcium chloride, and a parallel aide arm adapter fitted with a 250 m1. dropping funnel guaracd by a tub. con- taining anhydrous calcium chloride and a thermometer, was placed 4.2562 3. (0.02005 males) of g-urosyl benzaato in 75 m1. of sodium dried toluane. To thin cater mixture in added by means of the firepping funnel 30 m1. of the Grignard containing 0.758 milliaq./ml. or total of 22.7 millieq. or 0. £2? moles. the Grignard was added with agitation over a 10 minute pericd. The ether was removed, without concentrating the solution by adaing 19 50 ml. of sodium dried toluene to the reaction mixture and collecting the other in I Dean-Stark trap. after the other can removed, the system was refluxed for one hour at 110°C and the reaction quenched with 60 ml. of s 10% hydrochloric acid solution. After vigorously agitating the reaction mixture for 15 minutes it was transferred to s 250 ml. separatory funnel and the water layer discarded. The organic layer was washed twice with 100 ml. portions of distilled water, twice with 100 m1. portions of a 5% potassium hydrox- loe solution and three times with 100 m1. of distilled water. The reaction mixture was returned to the original reaction flask and the water remaining was szeotrOplcslly removed by the Dean-5tark trap. The reaction mixture was allowed to cool and transferred to s 250 ml. volumetric flask, made to volume with sodium dried toluene and then transferred to a glass stoopered bottle. All the other esters were caused to react using identi- cal conditions. The concentrations of the esters and Grignard used in these reactions are given in Table II. The yields of kctone and alcohol and the Amount of un- reacted ester obtained in each reaction will be given under Analysis of Products. Anallsis of Products The principal products in the reactions of mesityl- magnesium bromide with the aryl esters prepared are diaryl TABLE II INITIAL CGNCENTRATICNfi CF REAGENTS IN HEAC?IONS (jer- Mr- (~- CF hESITYLEAGNESIUfi BRoeIDE WITH flRYL Enigma Reaction Ester goncentretione Foleg/Litgg firignard Reagent Phenyl benzoats Phenyl grmethylbenzoate Phenyl.2-methylbenzoate Phenyl.g~ethylbenzoate Phenyl{grisoprOpylbenzoats ‘g-Cresyl benzoate ‘g-Cresyl‘g-methylbenzoate g-Cresyl‘2-methylbenzoate g-Cresyl‘g-ethylbensoate 'g-Cresyl‘g-ieooropylbenzoate 0.2000 0.2005 0.2000 0.2000 0.2010 0.2005 0.2004 0.2004 0.2001 0.8000 0.9525 0.2079 0.2625 0.2052 0.2181 0.2274 0.2274 .1. [Bat-W1.‘ "'7'? fiflaa-F __ .“ A' A e 9 . e . ‘ a ketones and triaryl tertiary alcohols. In all cases unre- acted ester remained at the end of tee reaction. These three organic compounds were determined by infrared analysis using a Perkin-Elmer model 21 Double Beam Infrared Epectrooho- tomster. The three standards used to determine these com- pounds wsre phenyl benzoate absorbing at 5.75 microns, benzephenone absorbing at 6.60 microns, and tripnenyl carbinol absorbing at 2.80 microns. Standard solutions of these substances dissolved in toluene were prepared end run on the Perkin-Elmer Infrared Spectrophotometer. Calibration curves were then prepared by plotting transmittancy versus concentration. These calibration curves were used for all of the other ketones, alcohols and esters prepared during this investigation. The Justification for using one member of toe series as the infrared analytical standard in determining the other members of the series hes been discussed by Bellamy (2?). To determine if there was any interference between ester, kctone and alcohol in the infrared absorption, a known solu- tion was made up containing 0.033 moles of phenyl benzoete, triphenyl carbinol and benzophenone. This solution was run on the Perkin-Elmer Infrared Spectrophotometer. The results found were 0.054 moles of the alcohol, 0.034 moles of the ketcne and 0.053 moles of the ester. These results indicate that there was no interaction among the three compounds. The infrared analysis of the ketone and ester were checked by volumetric methods using hydroxylamine hydro- chloride in pyridine for the ketone and hydrolysis with standard base in ethylene glycol for the ester. The results of the analysis of the reaction mixtures are given in Tables III and IV. Inrnéaso arisen Deternination of Ketones A calibration curve for ketones was made using the ab- sorption at 6.00 microns. Benzophenone was used as the standard on the Perkin-Elmer nodal 21 Infrared spectrOpho- tometer using 0.495 mm. sodium chloride solution cells. The eight solutions used for calibration were made up in sodium dried toluene. They were 0.0500, 00250, 0.0200, 0.0125, 0.0100, 0.0050, 0.0025 and 0.0010 moles per liter. The method of calibration was as follows: Both the reference and sample cells were filled with sodium dried toluene and the zero percent and hundred percent curves were made from 2 to 7 microns. Each of the eight solutions were then run scanning from 2 to 7 micron: and the peak height was found at 6.00 microns. To find the transmittancy, the peak height in one. was divided by the distance between zero and hundred percent lines in one. at 6.00 microns and these values plotted on graph paper to give concentration in moles per liter versus transmittancy. TABLE III INFRARED DETERmINATION 0F DIAREL KETONES, TRIARYL TERTIARY ALCOHULS AND UNREACTED AFYL ESTERS IN REACTIONS or uxsxrrugaawaszuu snowing WITH Esrxafi 23 ___‘ Reaction Ketono Tertiary Alcohol Unreacted (Moles) (Halon) Eater (Mole!) Phenyl benzoato 0.00570 0.00206 0.00765 Phenyl Ig-methylbonzoate 0.00450 0.00177 0.00940 Phenyl grothylbentoato 0.00428 0.002?0 0.01080 Phenyl - grisOprOpylbenzoato 0.00460 0.00220 0.01130 Phenyl ‘g-methylbenzoato 0.00698 0.00230 0.00680 .g-Crosyl benzoate 0.00420 0.00170 0.01040 g-Creayl ‘gymethylbenzoato 0.00270 0.00230 0.01140 l2¢0reay1 .g-ethylbenzoata 0.00670 0.00162 0.01360 ‘g-Cresyl grluoprOpylbenzoato 0.00550 0.00156 0.01450 g-Cre sy1 ‘E-mothylbenzoato 0.00500 0.00193 0.00880 s-.. o. 1. 24 TABLE IV DETERMINRTION OF DIARYL KETONES BY HYDROXYLAMINE HYDRO- CHLORIDE IN PYHIDIHE AHD OF UNREACTED AHYL ESTERS BY HYDROLISIS 0F ESTER WITH STANDARD BASE IN ETHYLENE GLYCOL FOR THE FOLLOWING REACTION NIXTURS A ‘- w.‘ Yv—v —. v Renotion Ketona Unreaoted (moles) Eater (molel) Phonyl.g-aoth11benxonto 0.00461 0.00957 Fhenyl‘grlsOpropylbenzonte 0.00470 0.01130 .970reny1 bonzoato 0.00407 0.01030 .g-Cronyl‘g-ethylbonzoate 0.00381 0.01300 u A _. o . _ - _ a n . v- i a I l o I ”A Table V gives the measured transmittancy for each of the standard solutions used. Figure 1 is the calibration curve obtained from the date of Table V. analysis of the ketones in the reaction mixture was made at two concentrations in each case. The first concen- tration was identical to that of the reaction mixture diluted to 250 ml. with toluene, the second concentration, half that or the first, was obtained by diluting s 5 m1. sample of the first concentration to 10 m1. using toluene as the diluent. The solution cells and general procedure used were the ease as those employed in the calibration studies. The results of the infrared analysis for ketones in each of the reaction mixtures are given in Table VI. Qetermination of Alcohols A calibration curve for alcohols was made using the ab- sorption at 2.80 microns. Triphenyl carblnol was used as the standard. The procedure was the same for the ketcnes ex- cept the solutions used were 0.1000, 0.0500, 0.0250, 0.020 , 0.0125, 0.0100, 0.0050 and 0.0025 moles per liter. Table VII gives the measured transmittency for each of the standard alcohol solutions and Figure 2 is the calibration curve ob- tained from the data of Table VII. Since the yield of alcohol was very small in all re- actions studied, it was necessary to concentrate the re- action mixture for alcohol analysis. This was done as 26 TABLE V INFRARED TRAEBfiITTANCY FOR STANDARD SGLUTIONS OF Beszcruznons IN TOLUENE Concentration of Trensmittenc ' Solution in Holes/liter y“ 0.0500 0.182 0.0250 0.366 0.0200 0.428 0.0125 0.587 0.0100 0.661 0.0050 0.814 0.0025 0.916 0.0010 0.978 “Obtained from Perkin-Elmer $0091 21 Infrared Spectroghotometer at 6.00 microns using 0.495 mm. sodium chloride solution cells. 27 Figure l . Standard curve of the ketone determined by means of a Perkin~Elmer, Model 21 Double Beam, Infrared Spectrophotometer using 0.495 mm. sodium chloride solution cells, with sodium dried toluene as the reference. Transmittancy Concentration (Moles/liter) -0.C4 40.05 0.02 1.00 L g l 0.90 0.80 ‘ 0.70 0.60 0.50 .40 #4 —II|‘I 03 .20 28 TABLE VI RESULTS 07 INFRARED DETERsEINATICNS 0F KETONEB IN REACTION MIXTURES Reaction Concentration of Yield of Mixture of lgetone (soles/liter) Ketone fl dethcd II method 2 Theoretical “A Phenyl benzoete 0.0057 0.0056 28.5 Phenyl 'grmethylbenzoate 0.0046 0.0044 25.0 Phenyl ,Eymethylbenzoete 0.0069 0.0065 54.5 Phenyl ‘gyethylbenzoate 0.0043 0.0041 21.5 Phenyl .g~isopr0pylbenzoate 0.0046 0.0045 25.0 ‘g-Creeyl benzoate 0.0043 0.0041 21.5 g-Oresyl ,g-methylbonzoate 0.0027 0.0027 15.5 ,gecreoyl 12-methylbenzoste 0.0050 0.0044 25.0 ,g-Cresyl ,gyethylbenzoate 0.0037 0.0005 13.5 ,gecresyl grieoprOpylbenzoete 0.0036 0.0056 17.7 a) method 1 concentration obtained from origi- nal 250 m1. of reaction mixture. b) Hethod 2 concentration obtained from di- luting 5 m1. of original to 10 ml. with toluene as diluent. .— ll TAELB VII IKFRAHZD TRANSkITTANCY FOR STRNLAED SQLUTIDfifl U? TRIPHENYL CARBINOL IN TOLUENE Concentration of Trancmittnnoy* Solution in Moles/liter 0.1000 0.540 0.0500 0.754 0.0250 0.651 0.0200 0.883 0.0125 0.924 0.0100 0.959 0.0050 0.980 0.0025 0.994 “Obtained from Perkin-Elmer Hodel 21 Infrared SpectrOphotometer at 2.80 microns using 0.495 mm. sodium chloride solution cells. C)! Q Figure 2 Standard curve of the alcohol determined by means of a Perkin-Elmer, Model 21 Double Beam, Infrared Spectrophotometer, using 0.495 mm. sodium chloride solution cells, with sodium dried toluene as the reference. Transmittancy Concentration (Moles/liter) ‘0 l L l ‘ .3, 0.20 1.00 0.90 0.00 0.70 0.60 0.50 0:50 0- O ”f .10 .09 .03 .02 .Ol follows: One hundred milliliters of the original 250 ml. reaction mixture was pipetted into a 250 ml. distillation flask and approximately 70 ml. of the toluene was distilled and discarded. The residue in the flask was diluted to 60 ml. in a volumetric flask with toluene, thus increasing the concentration by a factor of 2. After concentrating the solution, the infrared determinations were made by exactly the same procedure used for the ketones. The results of the e oohol determinations are given in Table VIII. Determination of Unreacted Ester A calibration curve for the esters was made using the absorption at 5.75 microns. Phenyl benzoate was used as the standard. The concentration of the solutions and the pro- cedure used were identical to that used for the ketones. Table IX gives the measured transmittancy for each of the standard solutions and Figure 5 is the calibration curve obtained from the data of Table IX. Two analyses for unreacted ester in each of the re- action mixtures were made exactly as for the ketone. The reason for this was that some of the esters were too concen- trated in the original 250 ml. of the reaction mixtures to run undiluted. The analyses were made and the results for the ester determination are given in Table X. f" {\3 TA LE VIII RESULTS OF INFRAPED DETERSIflATIONS - ALCCHGLB IN REACTION fiIXTURES f} '1' Reaction Concentration of -¥ield of Ketone Mixture of Alcohol (soles/liter) % Theoretical Phenyl benaoato 0.0021 10.6 Phenyl . ‘g-methylbenzoate 0.0018 9.0 Phenyl lg-methylbenzoate 0.0023 11.5 Phenyl . grethylbsnzoatc 0.0022 11.0 Phenyl ' .g-isOprOpylbonzoats 0.0022 11.0 .g-Cresyl benzoate 0.0017 8.5 g-Cruyl .g~methylbenzoate 0.0025 11.5 ‘ngrcsyl gymethylbenzoate 0.0019 9.5 ngrcsyl .g-ethylbenzoate 0.0015 7.5 ‘g-Crecyl ,g-iaoprOpylbsnzoate 0.0018 8.0 CA 03 T1131... ‘2 I )1 INFRARED TRANSIITTMJCY F-“Fk STIUEDI‘QI} SOLUTIONS OF PHTENYL BTQUATE IN TQLUEE‘JE Concentration of Transmittancy“ Solution in Moles/liter 0.0500 0.194 0.0250 0.328 0.0200 0.418 0.0125 0.549 0.0100 0.617 0.0050 0.785 0.0025 0.889 0.0010 0.951 *Obtained from Perkin-Elmer Model 21 Infrared SpectrOphotometer at 5.75 microns using 0.495 mm. sodium chloride solution cello. 64 Figure 5 Standard curve of the ester determined by means of a Perkin-Elmer, Model 21 Double Beam, Infrared Spectrophotometer, using 0.495 mm. sodium chloride solution cells, with sodium dried toluene as the reference. Concentration (Moles/liter) 0.05 0.04 0.05 -0.02 40.01 Transmittancy . . -.00 1.00 0:90 0:80 0.70 0.60 0.50 0.40 0.30 0.20 TABLE X 35 RESULTS 0? INFRARED DETEHEINATIONS PF TRE UNEEACTED ESTERS IN THE FEACTICN EIXTUEES —_' concentration of Iiold of Reaction Ester (moles/liter Unreacted Mixture of Eethod l‘fcetnod 2 Ester 1 Theoretical Phenyl benscctc 0.0077 0.0074 58.5 Phenyl gemstnylbensoatc 0.0095 0.0094 47.0 Phenyl gymcthylbcnzoate 0.0068 0.0057 54.0 Phenyl ,g-ethylbensoate 0.0108 0.0094 54.0 Phenyl .griscprcpylbenzoste 0.0115 0.0098 56.5 ,gycrcsyl bonsootc 0.0104 0.0099 52.0 lg-Orosyl .g-methylbenzoate 0.0114 0.0109 57.0 ‘g-Cresyl ‘gemethylbensoste 0.0087 0.0095 45.5 [gears-:1 gycthylbenzoate 0.0156 0.0124 68.0 3-Cresy1 ,g~isopropylbsnsoatc 0.0145 0.0150 72.6 a) Method 1 concentration obtained from origi- nal 250 ml. of reaction mixture. b) Method 2 concentration obtained from di- luting 5 ml. or ortginal to 10 ml. with toluene as diluent. 36 DETEEVINATION OF DIARYL KETCNLS BY HECROXYLAXINE HYL pfiv.u\ J33 HVTNCD For the purpose of checking the results obtained by infrared analysis, the oximoe of the dieryl ketonee were pre- pared by the method of Byrant and Smith (23) for the reaction mixtures of meeitylmegnesium bromide with phenyl‘g-methyl- benzoete, phenyl‘g—ieoprOpylbenzoete,‘g-creeyl benzoete and gfcreeyl.g-ethylbenzoate. A 50 m1. aliquot of the reaction mixture in 250 m1. toluene was pipetted into a Citrate of Magnesia bottle con- taining 25 ml. of 0.5 N hydroxyl amine hydrochloride in 80% ethanol and 100 m1. of a 20k pyridine in ethanol solution containing 0.02é bromOphenol blue indicator. The samples and two blenke containing only toluene with the reagents were beaten in a steam bath at 95°C for 4 hours. After heating, the samples and blanks were allowed to cool to room tempera- ture and the samples Were titrated to the color of the blank With 0.222 N potassium hydroxide in ethanol. Along with the unknown samples, two knowne were analyzed containing 0.00550 moles and 0.00249 melee of benzophenone. An example of the method of calculating the amount of ketone in the samples is given below for the o-creeyl benzoete reaction mixture: rpre a: baeP X 0““"911tv ml ;.91;3: : 0.00437 moles iUUU H 045 (\‘M No melee of xetone 37 For the two known eaeplee, reeulta Obtained were 0.00050 molee ketone for the known of 0.00360 aciee and 0.00249 moles for the Known of 0.00249 aolee. the reeulte obtained for the ketonee in the reaction mixtures by the hydroxyleninc hydro- chloride method nre given in table It end agree within ex- perimentnl error with thole obteined.by infrared analysis which ere given in Table III. .QETERXINATION C? THE UNREAGTFD AFYL E3TEFS BYMBYDRGLYfilg The mole: of unreacted ester were determined by hy- drolyeie in four of the reaction mixturee for the purpose of checking the results obtained by infrared analytic. The re- action Iixturec ueed were phenyl‘grmethylbennoate end phenyl [gel-aprOpylbenzoete endIg-creeyl benzoete end_g-creeyl 'g-ethylbenzonte. Hydrolyeie wee nocompliehed by pipetting n 00 m1. aliquot of the reaction mixture which nae been diluted to 250 m1. into a 250 m1. reflux fleck containing 25 ml. of a 0.518 N potoeeium hydroxide in 901 ethylene glycol 10% water aolution. A reflux condenser wee attached and the mix- ture wee caused to reflux for 24 hourel -The mixture wee then allowed to cool to room temperature and woe transferred to e 250 ll. beaker ueing 60 ml. of ethnnol on e weaning agent. Two blenke containing toluene and two known eclutiene con- taining 0.00000 end 0.00248 mole. of phenyl bentcnte in toluene were hydrolyzed nlong with the reaction nixturee. so The mixtures were titrated with 0.255 N hydrochloric acid in ethanol employing a Beckmen Rodel 0 pH meter equipped with a glass and a calomel electrode. The equivalence point of the blank, known and unknown mixture was found by plot- ting pH vereue milliliters of hydrochloric acid added. The difference in the titration between the blank and known or unknown mixture was the measure of the milliequivalente of the aryl acid and hence equal to the nilliequivelente of the unreacted nryl ester. An example of the method of calcu- lating the melee of unreacted ester is given below for the ‘g-creeyl benzoete reaction mixture. melee of unreected ester»: _Lml. x H for blank ml. 2 N for Samgle) 50/{2’1 Du X 1000 molee of.g-creeyl benzoete : 0.2 x 100 0.01050 moles b For the two known samples, results obtained were 0.00349 moles enter for the known of 0.00350 molee and 0.0024? molee ester for the known of 0.00248 moles. The reeulte obtained for the unreected eetere in the reaction mixturee by hydroly- eie are given in Table IV and agree within experimentel error with those obtained by infrared analysis which are given in Table 111. "; 3-4- ESULTS The results of this study will be presented under four headings: l) the percent of ester reacting: 2) the percent of ketone complex‘ formed, calculated from the yield of ketone formed plus the yield of alcohol formed; 3) the per- cent of this ketone complex reacting to form alcohol; and 4) the percent of Grignard reagent reacting, calculated from the yield of ketone formed plus twice the yield of alcohol formed. (It takes two molecules of Grignard reagent to form one molecule of alcohol.) In general, the results may be summarized as follows: The percent of eater reacting appears to decrease with increasing steric hindrance in the esters, while the other factors, (percent of ketone complex formed, percent of ketone complex reacting to form the tertiary alcohol and percent of Grignsrd reagent reacting) are relatively in- sensitive to the steric changes in the sroyl portion of the ester. Significant changes in these factors appear with substitution in the ortho position of the eryloxy portion of vTheketone complexfiis oefined as the intermediate resulting when one molecule of Grignerd reagent reacts with one mole- cule of enter. This complex may decompose to give the ke- tone, it may react with more Grignard reagent to give the tertiary alcohol or it may remain intact until hydrolysis at which time it will decompose to give the ketone. A - 9 x \ s ‘ , y I I . ' 1 '- o . -‘ ‘ ‘ u- . 4 . l O 40 the ester. The results are summarized in Table XI and will be discussed in more detail in the paragraphs which follow. Percent Ester Reacting A marked decrease in the percent of ester reacting is observed when sterio hindrance in either the eroyl or aryloxy portion of the ester is increased. The magnitude of this decrease can be seen by comparing the values obtained from the phenyl benzoete reaction with that from the phenyl grisoprOpylbenzoste reaction, 61.7% in the former against 43.5% in the latter and by comparing the values obtained from the phenyl.g¢ethylbenzoate reaction with that from the 'g-cresyl.g-ethylbenzoate reaction, 46% on the former compared to 32% in the letter. Both phenyl and g-cresyl‘g-methyl- benzoste reacted more rapidly with the Grignsrd reagent then their corresponding phenyl and gecresyl benzoate analogues. fiercent Ketone Comolex Formgi The results are discussed unéer the heading Percent Ketone Complex Formed rather than under Percent Ketone Formed because presumably each molecule of alcohol produced consumes one molecule of the ketcne or ketone complex. The total moles of ketone or ketone complex produced is measured by adding the number of moles of alcohol and ketone. The values for the percent ketone complex formed were obtained by di- viding the total moles of ketone cr ketone complex produced v ‘ _, . , i u... k .. -. . . ' . . . . .q, TABLE XI 41 PERCENT ESTER REACTINO, PERCENT KEToNE sceptex roses , rsscswr KETONE onePLex convsswsn co ALCGNOL AND pnscssr csxcnsno REACTING IN THE REACTICNS or MEBITYL- sAGNESIUh Ester X Enter f Ketone BROeIDE WITH THE ARYL ESTEEB w , , .7” fl Ketone Complex 1 Grignerd Reacting Complex Converted Reacting Phenyl benzoate Phenyl ‘gynethylbcnzoate Phenyl ‘gyethylbenxcste Phenyl grisopropylbenzcate Phonyl gruethylbenzoate ‘g~Cresy1 benzonte geareeyl ‘g-methylbenzoete grcreeyl ‘g-ethylbonzeate 2.61-er1 griIOpropylbenzoate g-Cresyl ‘Eynethylbenaoate 61.7 53.0 46.0 43.3 66.0 48.0 43.0 32.0 27.5 66.0 38.8 31.4 32.4 34.0~ 46.4 29.5 28.0 26.1 Formed To Alcohol 26.3 28.2 24.7 28.8 46.0 29.1 49.1 40.2 43.4 45.0 57.9 38.0 '- - - b - > < - . -. - ... _ ... yc— ‘ v v . < ' 1’. .Jd “ —" ‘ . ’ . L. . - _ -- ‘w ~ ~ I ' ‘ ' I u— ‘ I . . < . I ,‘ ,. . 4- . ‘ _ . , y. .— .. . ' ‘ ' ’ ' . I. u . , -. . - - ., _ - i - - ' J - . a ‘ ‘ . N -- ‘ ' A, 7 ‘ \ .- .-- . Y Q 1 . .. I I ~ _.. . _- , fl . ' " n! .. —2 by the maximum number of moles of kstons which could poten- tinlly be formed. Table XI indicates that the percent of ketone complex formed is virtually independent or the size of the crthc sroyl substituent provided this substituent is not a hydro. gen stom. Compsring the values obtained for the phenyl with thoss.tcr ths‘grcrssyl ester, it can be seen that there is a significant decrease in reactivity of the gearesyl compounds over their phenyl analogues. The magnitude of this decrease can be visuslised by comparing the values obtained for the phenyl g—isoprcpylbenscats reaction with that for the grcrssyl.gyiscpropylbensoste reaction, 34% in the former against 25.8fl in the latter. Here, as in the case of percent esters reacting, the results indicate that the E-Iethyl- benzoatss are more reactive than the other esters in the formation of the kstone complex. .Eercent KetonefiConplex Reacting to Form the Alcohol To form n molecule of alcohol, one molecule of ketcne or ketone complex must be consumed. The percent or ketone complex reacting to form alcohol is round by dividing the number of moles of alcohol produced by the total number of moles of ketcne complex formed. Table XI indicates that the percent of ketone complex reacting to form alcohol is not appreciably effected by the orthc sroyl substituent, provided the eubstituent is not a A [r f i ‘ . _. I .4 . o , V V ' - t ' A - u'n . . . -. . . n- ‘ ~ ' N ,l e . O . . , : ( ,- . . " "“ v7: - -.._.’ 43 hydrogen atom. There is a slight increase in the percent of kstcne complex reacting when comparing crtho hydrogen to crtho alkyl substituents. This may be interpreted as a loosening of the bonds or the complex by the ortho alkyl sub- stituant, or the electron density at the reacting center is increased by the ortho alkyl substitusnt. The magnitude of this increase can be seen shen comparing the values obtained from the phenyl benzcate reaction to that from the phenyl lg-ethylbenzoats reaction, 25.63 in foraer against 33.95 in the latter. The same trends are noted in comparing the ortho substitusnts in the gecresyl esters to g-cresyl bensoate. But upon comparing the phenyl and.g-cresyl esters as to the percent ketons couple: reacting, it is round that the ortho aryloxy substituent does not appreciably effect the reactiv- ity, except in the case of the‘Q-cresyl‘g-aethylbensoats which is unexplainable. The results indicate that the con- plex formed from the phenyl and‘g-cresyl.p-nethylbenzcates are less reactive than their corresponding phenyl and 23erssyl banzoats analogues and less complex is converted to alcohol. .Egrcent Grignerd Reagent Reacting The percent of Grignard reagent reacting is found by adding the moles of ketone formed to twice the moles of alcohol formed and dividing this sum by total moles of Grignard that could react. The Grignerd is the limiting rector because it is in equal molar proportion to the esters 44 and it has two methods of reacting. It reacts vith the enter to form the complex and can then react with the com- plex to form the alcohol. The results in Table XI indicate that the percent Grignard reacting is virtually independent of the size of the ortho aroyl substituent in both the phenyl and grcresyl esters. There is a decrease in the percent Grignerd re- acting when comparing crtho hydrogen to ortho alkyl sub- stitution. The magnitude or this decrease can be seen by comparing the values obtained from the phenyl bensoate re- action with that from phenyl gynethylbenzoate reaction, 49.1% in former against 40.2} in letter. There is also a decrease in reactivity when comparing phenyl caters to g-cresyl esters. This can be seen by comparing the values obtained from phenyl‘g-isopropylbensoate reaction to that from gycresyl.g-isoprcpylbenzoate reaction, t5.0% in former against 35.6% in letter. The results indicate that the Grignaré reagent is more reactive toward the ponethylbenzoatee than their correspond- ins gralkyl benzoate analogues. .. , . . . . . l . u v . f. . .. v 's O a _ v , . y 4. i . 4 . . a... . . o . . .q 2. s. . a . .l. a e - . . O . 1 . . . . . x e . . a . . . . .. . . . , . r e . .. . . i n g « . _ . A . .5 . . . . . . a , . . : ,I . ._ A . \ .. _ M . . a . .. l v 4 . : \ u r o . , . . ~ - . f , _ .c , . . s e i A w l , -4 1.: . . . . W n i l s u s C e \ x . . . .s . . 48 4) Toluene was used as a reaction medium for two rea- sons: a) to avoid the complex formation which occurs between the Grignard reagent and others. This complex would intro- duoe an uncontrolled sterio factor; b) to acquire the higher reaction temperatures obtainable by using toluene as the solvent. 11. Table XII summarizes the results as relative reactiv- ities compared to phenyl benzoate. The table shows that the relative reactivities of the Grignard reagent, the ester, and the ketone complex are remarkably constant for the ortho substituted phenyl esters. Comparing the results obtained for the phenyl esters with those of the o’cresyl esters indi- cates that the gecresyl esters are less reactive by about 20%. But the ketone complex formed from the g—cresyl esters apparently reacts more readily with the Grignard reagent than the complex formed from the phenyl esters. This may be due to the loosening of the ketone complex by a sterio effect or an electrical effect of the g—methyl substituent of the cresyl group. The electron donating effect of this 2-methyl group would increase the basicity of the sryloxy oxygen atom of the ester. This should favor a greater reactivity of this oxygen atom toward the magnesium atom of the Grignard re- agent. The sterio effect due to the1g-methyl group in the aryloxy portion of the ester causes the bonds of the complex to be lengthened thereby making it easier for the second 'n ‘v 46 4) Toluene was used as s reaction medium for two rea- sons: s) to svoid the complex formation which occurs between the Grignsrd reagent snd sthers. This complex would intro- duos an uncontrolled sterio factor; b) to acquire the higher reaction temperstures obtainable by using toluene as the solvent. 11. Table XII summarises the results as relstive reactiv- ities compared to phenyl benzcsts. The table shows that the relstive reactivities of the Grignsrd reagent, the ester, and the ketone complex are remarkably constant for the ortho substituted phenyl esters. Comparing the results obtained for the phenyl esters with those of the.g-cresy1 esters indi- cates that the gecresyl esters are less reactive by about 20%. But the ketcne complex formed from the peeresyl esters apparently reacts more readily with the Grignard reagent than the complex formed from the phenyl esters. This may be due to the loosening of the ketone complex by s sterio effect or en electrical effect of the g~methyl substituent of the cresyl group. The electron donating effect of this g-methyl group would increase the besicity of the srylcxy oxygen atom of the ester. This should favor a greater reactivity of this Oxygen atom toward the magnesium atom of the Grignard re- agent. The sterio effect due to the gfmethyl group in the aryloxy portion of the ester causes the bonds of the complex to be lengthened thereby making it easier for the second 47 TABLE XII THE REACTIVITIES C? KSTBR, KETCNE CCKPLEX AfiD GRIGNARD, RELATIVE TO PHVNYL BCHZOATE f ‘_ —— Reactivity Feectivity Reactivity of Ester“ of Xetone of Grignerd Complex*' 1 1 1 Benzoete 0.759 1.086 0.774 0.808 1.065 0.819 lg-fiethyl- benzoste 0.644 1.764 0.744. 0.834 1.280 0.886 bensoste 0.672 1.100 0.686 0.876 1.220 0.916 gelsopropylo bensoets 0.665 1.140 0.684 1.191 0.934 1.179 3-1.1ethyl- bsnsosts 0.891 1.052 0.902 Phenyl g-Cresyl ’Meesured by the amount of ketone complex formed. “*Eeesured by the amount of alcohol formed. ‘Q‘.- 48 solecule of the Grignard reagent to attack the aryl carbon. This would then csuse the bond between the eryl carbon and the arylosy oxygen atoms to be broken and would remove this sterio factor. Boss interesting comparisons can be made in the relative reactivities of the esters in these reactions. One interesting comparison is that of phenyl bensoste, phenyl gynethylbenscete and‘g-cresyl benscsts. From Table x11, it is observed tbst sn,gyoethy1 substitusnt on the aroyl portion of the ester decreases the reactivity of the ester by about 20;. When sn‘g-nethyl group is substituted on the srylcxy portion of the ester, the reactivity is decreased by about 25} as compared with the unsubstituted ester and 60% so compared with pheny1,grnethylbenzcste. this would indi- cate that substitution on the sryloxy portion of the ester has s greater sterio effect than substitution on the eroyl portion. Exactly the same trend is found for the reactivity of Grignsrd reagent. Considering the ketone complex, it is found that the relative reactivities sre approximately equal and that en .gemsthyl substitution on either eroyl or sryloxy portion of the ester has the seas effect. The comparison of the relative reactivities of the Phinrl‘gelethvle snd,2elethylbenzostes with phenyl and .grsrssyl bsnsoates is also of interest. From Table XII, it 1., 49 is seen that gynethyl substitution on the aroyl portion de. creases the reactivity of the ester. On the other hand, gynethyl substitution shoes an increase in reactivity of about 201 compared to the unsubstituted bsnacates and an in- crease of 405 compared with the.genethylbenzoates. _Like- sise, a 45; increase in reactivity is observed.vhen compar- ing phenyl‘p-msthylbsnaoatc to.g-cresyl benzoate. These increases in reactivity could be due to the elec- tron donating effect of the‘g-msthyl group, which makes the sroyl oxygen atom of the ester more basic and more reactive toward the magnesium atom of the Grignard reagent. .Deter- ainations of the amount of Grignard reagent reacting support these observations. Considering now the reactivity of the ketone complex, a decrease in reactivity is observed for the phenyllp-nethylbcnecats compared to phenyl bensoate and phenyl'gemethylbenzoate by about 10%. It was observed that ths‘g-Iethyl group enhances the formation of the complex by its electron donating effects. But this sane effect seems to be a detriasnt to the reactivity of the complex. This would suggest that the reactivity of the complex is governed more by sterio factors than by electrical factors. Another interesting comparison is of phenyl and‘gyoresyl ‘grasthylbenseatss eith phenyl and‘goeresyl‘p-aethylbenzoates. It was observed that the phenyl‘ponethylbenscate was more reactive toward the Grignard reagent than phenyl . 4 C .. .- . . . ‘ v r ' ‘ ' ’ I v ‘- ‘gemcthylbcnsoate. on the other hand, the reactivities of the respective hetone complex was reversed. Considering the re- activities of the phenyl esters compared to the g-cresyl esters, it is observed that the reactivity of the ester and of the Grignard reagent decrease going from phenyl to g—cresyl. The largest decrease is observed for the transi- tion from phenyl.p-methylbsnzoate to g—cresyl p-methylbenzoate. The magnitude of this decrease is about 30%. The reactiv- ities for the transition from phenyl.2-methyl- to gycreeyl .g-methylbsnzoate is decreased about 16%. On the other hand, the reverse of these transitions is observed for the re- activity of the ketons complex. ‘The transition of phenyl g-nethyl- to‘ggoresylIg-mcthylbenzoate has an increase of re- activity by about 70%, whereas, the transition of phenyl ‘pymethylo to gyoresyl prmethylbenzoate has an increase of only 103. This would further indicate that the gemsthyl group of the aryloxy portion of the ester has the greatest effect upon reactivity and that this effect is more sterio than electrical in nature. Also, there must be some electrical effects of the lgcncthyl substituent because the reactivity of the ketone complex, while it increases from phenyl to g~cresyl, is not as large an increase as for the g-methyl substituted esters. This electrical effect which increases the formation of the complex must also form a more stable complex by forming a Va v.4 ‘ . A . a . k I. .1 .i. L. a . l, . uh. I. 2 y _ H n. A .rt . . Q , t v 1 .. . L to. .. O '0 . ‘ 51 charge cloud around the functional group of the complex which repels the second molecule of attacking Grignard reagent. Likewise, if the effect caused by.g-methyl substitutent of aryloxy group (g-cresyl) was electrical rather than sterio, the electron donating effects of the methyl group should make the reactivity of the‘groresyl‘prmethylbenzoate at least as reactive as the‘gyoresyl‘grnethylbenzoate, but this is not observed. The reactivity of the complex of the groresyl ,p-methylbensoate is comparable to that of gycresyl benzoate and, therefore, suggests that the increased reactivity of the-g—cresyl.g-msthylbensoate compared to the‘g-oresyl 'grmethylbenzoste is due only to sterio rather than electri- cal factors. Ths effects of further increasing the size of the gyalkyl substituent (on aroyl portion) does not appreciably change the reactivity of the ester compared to the.g-methy1 substituent, except in the case of the reactivity of the complex formed. This difference in reactivity of the com- plex is interpreted as due to the larger alkyl group causing more loosening of the complex as compared to the g—methyl group and thereby making it easier for the second molecule of Grisnard reagent to attack the complex formed from the phenyl esters. In the case of the,g-cresyl esters, this in- creased steric factor of alkyl groups on both the sryl and aryloxy portions hinder further attack by the highly hindered Grignard, even though the complex may be loosened, the alkyl groups shield the acyl carbon from any further attack. .Tfflqw m ”I. ‘2‘ 1. 55 COHGLUBIQHQ Results of this study indicate that bulky groups in the two sterio areas studied, i.e. the aroyl portion of the ester and the arylcxy portion of the ester, hinder the reaction of a highly hindered Grignard reagent with esters. The degree of hindrance has been shown to be virtually independent of the size of the ortho substituent due to the large degree of steric hindrance in the Grignard re- agent. Formation of the initial complex of the Grignard and cater has been ehcvn to be virtually independent of sterio factors of the order of magnitude employed in thlfl study. The formation of alcohol has been shown to be greatly influenced by sterio factors in the aryloxy portion of the ester. (3) (4) (5) (6) (7) (8) (10) (11) (12) (is) (14) (15) BIBLIOGRAPHY D. R. Boyd and H. H. Matt, J. Am. Chem. 300., g9, 898-910 (1927). E. R. Kohler and R. Baltzly, J. Am. Chem. 800., pg, L. Smith and C. Guss, J. Am. Chem. Soc.,,§§, 804-805 (1957). R. T. Arnold, H. Blank, and B. Liggett, J. Am. Chem. coo.,,gg, 3444-5446 (1941). R. C. Pusan, E. a. Bottorff, and S. B. Speck, J. Am. Chem. Soo.,,§g, 1239-1240 (1942). F. C. Whitmore and R. 8. George, J. Am. Chem. Soc., pg, 0. R. Heuser, P. Saperstein and J. C. Shivers, J. Am. Chem. Soc.,,zg, 606-608 (1948). W. Triebs; (H. S. Newman, Steric Effects in Organic Chemistry, New York, Kev Xork, John Wiley a Sons, Inc., 1955, pp. 406-417); Ann. 656, pp. 10-22 (1944). F. C. Whitmore and C. E. Lewis J. Am. Chem. Soo.,,Qg, 1616 (1942) and,§3, 2964 (1942 . F. L. Greenwood, F. C. Whitmore and H. M. Crooks, J. Am. Chem. Soc.,,§Q, 2028 (1938). F. C. Whitmore at al, J. Am. Chem. Soo.,,§§, 643 (1941). F. C. thdtmore and L. P. Block, J. Am. Chem. 500.,‘gé, l619 (1942). I. O. Whit-ore and C. T. Lester, J. Am. Chem. Soc.,,§g, 1247 (1942). F. C. Whitmore and D. I. Randall, J. Am. Chem. Soc.,,§g, 1242 (1942). l. S. Kharasch and 0. Feinmuth, Grignard Reactions of Nonmetallic Substances, New York, Prentice Hall, 1954, pp. 666-664. _ x L. I § D ' w I a a s ' . ( I ‘ Q n ‘ I s ‘ ‘ I e . . I I 0 . ‘ v u ' . s 1 . p 1‘ s u ' v - , . . e g ‘ ( I x _ . s 1 . . A $ ' I J ‘ I \ ‘ t \ ‘ ‘ ‘ V . u . , a , l . 7 . l s O - A t . t a-.. (16) (17) (18) 55 Arthur J. Pastor, A Study of Steric Effects in the Pe- actions of Aryl Esters With Aryl Grignard Reagent, haster's Thesis; Michigan State University, E. Lansing, aiehigan, (1957). Organic Synthesis Collected Volume 11, John Wiley a Sons, lnc., London and New York, (1946) p. 95. Organic Synthesis Collected Volume I, John Wiley a Sons, Inc., London and New York, 2nd Edition, (1946) p. 22 . M. Crawford and F. H. C. Stewart, J. Am. Chem. 800., pp. 4445-4447 (1952). Barerio Zuffanti, J. Chem. Education, 2;, 481 (1948). W. J. Hickenbottom, Reactions of Organic Compounds, Longmons, Green and Co., Inc., New York, New York, End Edition, (1950) pp. 98-99. L. Bellamy, The Infra-red Spectra of Complex molecules, John Wiley and Sons, Inc., London and New York, (1954) pp. 129-136. a. E. D. Bryant and D. M. Smith, J. Am. Chem. 300., ,§1, 57-61 (1955). .00 ———e a; 90+ 80- .70. 1», A O DUI-g u p i «H e 2 .50.w | S. e« .40. 30. n '60; , Figure 4 Infrared spectrogram of the reaction mixture of 0.02 moles of mesitylmagnesium bromide With 0.02 moles of phenyl 2-isopropylbenzoate .10. dissolved in 250 m1. of sodium dried toluene, determined by means of a Perkin-Elmer Model 21 Double Beam, Infrared Spectrophotometer, using 0.495 mm. sodium chloride solution cells 00 With sodium dried toluene as the reference. 2 s 4 5 6 '7 a Wavelength (Microns) ‘ I \ 57 .00 ‘\nF1v/fl\_—\J’/~\—/r: .90. 90. I .70. S 60)- C a p p H a CD .50. f; a E .40. .50. .20. Figure 5 Infrared spectrogram of the reaction mixture of 0.02 moles of mesitylmagnesium bromide with 0.02 moles of phenyl 2—methy1benzoate .10. dissolved in 250 ml. of sodium dried toluene, determined by means of a Perkin-Elmer Model 21 Double Beam, Infrared SpectrOphotometer, using 0.495 mm. sodium chloride solution cells 00 With sodium dried toluene as the reference.‘ 2 3 '4 5 6 '7 8 Wavelength (Microns) ’.90. (D O I .70. .50. .50. .40. .50. .10. Transmittancy 58 Figure 6 Infrared spectrogram of the reaction mixture of 0.02 moles of mesitylmagnesium bromide With 0.02 moles of o-cresyl g-ethylbenzoate dissolved in 250 ml? of sodium dried toluene, determined by means of a Perkin-Elmer, Model 21 Double Beam, Infrared Spectrophotometer, using 0.495 mm. sodium chloride solution cells With sodium dried toluene as the reference. A A L A b .00 5 6 7 8 3 4 Wavelength (Microns) \AU 59 ,0 - *rLJ 90. 1 .so. Ff,“ .70. 60. ‘2; C GS p p H e .50. 2 1 m a B I .40. ‘ .30? Figure 7 20 Infrared spectrogram of the reaction mixture “ of 0.02 moles of mesitylmagnesium bromide With 0.02 moles of phenyl -methylbenzoate dissolved in 250 ml. of so ium dried toluene, . determined by means of a Perkin-Elmer, Model .10, 21 Double Beam, Infrared Spectrophotometer, using 0.495 mm. sodium chloride solution cells with sodium dried toluene as the reference. .00 . 1 , J , 4 . 9 3 4 5 5 7 Wavelength (Microns) ‘1Lrltli‘l ‘ \ 1.00 0.50. 0.40. 0.50. W “VJ. Transmittancy . Figure 8 Infrared spectrogram of the reaction mixture of 0.02 moles of mesitylmagnesium bromide with 0.02 moles of g-cresyl benzoate dissolved in 250 ml. of sodium dried toluene, determined by means of a Perkin-Elmer Model 21_Double Beam, Infrared Spectrophotometer, using 0.495 mm. sodium chloride solution cells with sodium dried toluene as the reference. 1 l J l 1 l 0.00 3 4 5 6 7 . 8 Wavelength (Microns) cmtmmv rmmDate Due Demco-293 ._____ .__ MICHIGAN STATE UNIVERSITY OF AGRICULTURE AND APPLIED SCIENCE DEPARTMENT OF CHEMISTRY EAST LANSING, MICHIGAN (WY LIBBAPYOW a. Thusis 1957 McComb, Robert 3. Astudy of the steric effects in thereaetion of aryl esters with an aryl grignard reagent. {C all-52.35 1'1; ..’ «1:5..- LJ'. 1‘“ "'IIIIIIIIIIIILIIIIIIIIIIIIIIII”